Re: Bk6.6Ch6.9 glibc `__stack_chk_guard' work around.

[Chris Staub]

On 03/31/2010 04:06 PM, Szabolcs Gyalókay wrote:
> Hi All,
>
>
> OK, then libssp is not supposed to be linked, then why isn't somebody
> asking the obvious question?
>
> In the original post Pete wrote:
>
> It seems that libssp is suppose to be pulled in by
> the gcc option '-fstack-protector' so that '-lssp'
> shouldn't be necessary on the link line.
>
>
> So libssp is included because of '-fstack-protector'. I assume
> '-fstack-protector' is also not supposed to be there. Then why is it here:
>
> gcc res_hconf.c -c -std=gnu99 -fgnu89-inline -O2 -Wall -Winline
> -Wwrite-strings -fmerge-all-constants -g -Wstrict-prototypes
> -DIS_IN_nscd=1 -D_FORTIFY_SOURCE=2 -fpie -fstack-protector
>
> I can only say for sure: I'm pasting instructions from the book, and I
> did not put it there.

I also see -fstack-protector on my Glibc build, but it doesn't try to 
pull in libssp. Perhaps that isn't the real problem.
>
> I think you have to start entertaining the possibility that something is
> wrong with the book.

Not likely when many other people have built it and have not had the 
same problem. The only other possibility I can think of is a possible 
problem with some certain hosts, but that should no longer be an issue 
once inside chroot.

> As an addition, the output of glibc's configure:
>
> checking for -fno-toplevel-reorder -fno-section-anchors... yes
> checking for -fstack-protector... yes
> checking for -fgnu89-inline... yes

Yeah, I see that too, and again, my Glibc build does not attempt to link 
to libssp.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Bk6.6Ch6.9 glibc `__stack_chk_guard' work around.

[Chris Staub]

On 03/31/2010 04:22 PM, Chris Staub wrote:
> On 03/31/2010 04:06 PM, Szabolcs Gyalókay wrote:
>> Hi All,
>>
>>
>> OK, then libssp is not supposed to be linked, then why isn't somebody
>> asking the obvious question?
>>
>> In the original post Pete wrote:
>>
>>  It seems that libssp is suppose to be pulled in by
>>  the gcc option '-fstack-protector' so that '-lssp'
>>  shouldn't be necessary on the link line.
>>
>>
>> So libssp is included because of '-fstack-protector'. I assume
>> '-fstack-protector' is also not supposed to be there. Then why is it here:
>>
>> gcc res_hconf.c -c -std=gnu99 -fgnu89-inline -O2 -Wall -Winline
>> -Wwrite-strings -fmerge-all-constants -g -Wstrict-prototypes
>> -DIS_IN_nscd=1 -D_FORTIFY_SOURCE=2 -fpie -fstack-protector
>>
>> I can only say for sure: I'm pasting instructions from the book, and I
>> did not put it there.
>
> I also see -fstack-protector on my Glibc build, but it doesn't try to
> pull in libssp. Perhaps that isn't the real problem.

Hmm, sorry I just noticed that my latest Glibc log is from an older 
version. Maybe I need to look at the current one again...
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Bk6.6Ch6.9 glibc `__stack_chk_guard' work around.

[Chris Staub]

On 03/31/2010 04:33 PM, Bruce Dubbs wrote:
> Szabolcs Gyalókay wrote:
>>
>> gcc res_hconf.c -c -std=gnu99 -fgnu89-inline -O2 -Wall -Winline
>> -Wwrite-strings -fmerge-all-constants -g -Wstrict-prototypes
>
>> -DIS_IN_nscd=1 -D_FORTIFY_SOURCE=2
>> -fpie
>> -fstack-protector
>> -DNOT_IN_libc=1
>
>> -o /sources/glibc-build/nscd/res_hconf.o -MD -MP -MF
>> /sources/glibc-build/nscd/res_hconf.o.dt -MT
>
> Removing non-relevant stuff and reformatting a bit, my reference build has:
>
> gcc res_hconf.c -c -std=gnu99 -fgnu89-inline -O2 -Wall -Winline
> -Wwrite-strings -fmerge-all-constants -g -Wno-strict-prototypes
>
> -Wno-write-strings
>
> -Wstrict-prototypes
>
> -fexceptions
>
> -I../include -D_LIBC_REENTRANT -include ../include/libc-symbols.h
> -Dgethostbyname=res_gethostbyname -Dgethostbyname2=res_gethostbyname2
> -Dgethostbyaddr=res_gethostbyaddr -Dgetnetbyname=res_getnetbyname
> -Dgetnetbyaddr=res_getnetbyaddr -o
> /sources/glibc-build/resolv/res_hconf.o -MD -MP -MF
> /sources/glibc-build/resolv/res_hconf.o.dt -MT
> /sources/glibc-build/resolv/res_hconf.o
>
> It would seem that your Makefile is significantly different from mine.
> That would indicate to me that there is something wrong with your
> Chapter 5 build as once you get into chroot, the builds really should be
> the same.
>
> -- Bruce

OK, I just double-checked with current LFS dev, and I've got...

configure
...
checking for -fno-toplevel-reorder -fno-section-anchors... yes
checking for -fstack-protector... yes
checking for -fgnu89-inline... yes
...

And of course Glibc builds just fine without complaining about libssp.

Also, it seems you're confusing 2 different parts of the build. I've got 
res_hconf.c being compiled twice - once in the "resolv" dir and once in 
"nscd". Note that the previous user's paste is from the "nscd" dir and 
yours is from "resolv", which explains the differences. As for me, in 
the "resolv" dir...

...
gcc res_hconf.c -c -std=gnu99 -fgnu89-inline -O2 -O3 -Wall -Winline 
-Wwrite-strings -fmerge-all-constants -g -march=i486 -mtune=native -pipe 
-Wno-strict-prototypes -Wno-write-strings -Wstrict-prototypes 
-mpreferred-stack-boundary=2  -fPIC -fexceptions   -I../include 
-I/home/root/build/glibc-build/resolv -I/home/root/build/glibc-build 
-I../sysdeps/i386/elf -I../nptl/sysdeps/unix/sysv/linux/i386/i68
6 -I../sysdeps/unix/sysv/linux/i386/i686 
-I../nptl/sysdeps/unix/sysv/linux/i386
-I../sysdeps/unix/sysv/linux/i386 -I../nptl/sysdeps/unix/sysv/linux 
-I../nptl/sysdeps/pthread -I../sysdeps/pthread 
-I../sysdeps/unix/sysv/linux -I../sysdeps/gnu
  -I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet 
-I../sysdeps/unix/sysv/i386 -I../nptl/sysdeps/unix/sysv 
-I../sysdeps/unix/sysv -I../sysdeps/unix/i386 -I../nptl/sysdeps/unix 
-I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/i386/i686/fpu 
-I../nptl/sysdeps/i386/i686 -I../sysdeps/i386/i686 
-I../sysdeps/i386/i486 -I../nptl/sysdeps/i386/i486 -I../sysdeps/i386/fpu 
-I../nptl/sysdeps/i386 -I../sysdeps/i386 -I../sysdeps/wordsize-32 
-I../sysdeps/ieee754/ldbl-96
-I../sysdeps/ieee754/dbl-64 -I../sysdeps/ieee754/flt-32 
-I../sysdeps/ieee754 -I../sysdeps/generic/elf -I../sysdeps/generic 
-I../nptl  -I.. -I../libio -I.  -D_LIBC_REENTRANT -include 
../include/libc-symbols.h  -DPIC -DSHARED 
-Dgethostbyname=res_gethostbyname -Dgethostbyname2=res_gethostbyname2 
-Dgethostbyaddr=res_gethostbyaddr -Dgetnetbyname=res_getnetbyname 
-Dgetnetbyaddr=res_getnetbyaddr -o /
home/root/build/glibc-build/resolv/res_hconf.os -MD -MP -MF 
/home/root/build/glibc-build/resolv/res_hconf.os.dt -MT 
/home/root/build/glibc-build/resolv/res_hconf.os
...

And in "nscd"...

...
gcc res_hconf.c -c -std=gnu99 -fgnu89-inline -O2 -O3 -Wall -Winline 
-Wwrite-strings -fmerge-all-constants -g -march=i486 -mtune=native -pipe 
-Wstrict-prototypes -mpreferred-stack-boundary=2  -DIS_IN_nscd=1 
-D_FORTIFY_SOURCE=2 -fpie -fstack-protector   -I../include 
-I/home/root/build/glibc-build/nscd -I/home/root/build/glibc-build 
-I../sysdeps/i386/elf -I../nptl/sysdeps/unix/sysv/linux/i386/i686 
-I../sysdeps/unix/sysv/linux/i386/i686 
-I../nptl/sysdeps/unix/sysv/linux/i386 -I../sysdeps/unix/sysv/linux/i386 
-I../nptl/sysdeps/unix/sysv/linux -I../nptl/sysdeps/pthread 
-I../sysdeps/pthread -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu 
-I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet 
-I../sysdeps/unix/sysv/i386 -I../nptl/sysdeps/unix/sysv 
-I../sysdeps/unix/sysv -I../sysdeps/unix/i386 -I../nptl/sysdeps/unix 
-I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/i386/i686/fpu 
-I../nptl/sysdeps/i386/i686 -I../sysdeps/i386/i686 
-I../sysdeps/i386/i486 -I../nptl/sysdeps/i386/i486 -I../sysdeps/i386/fpu 
-I../nptl/sysdeps/i386 -I../sysdeps/i386 -I../sysdeps/wordsize-32 
-I../sysdeps/ieee754/ldbl-96 -I../sysdeps/ieee754/dbl-64 
-I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754 
-I../sysdeps/generic/elf -I../sysdeps/generic -I../nptl  -I.. -I../libio 
-I.  -D_LIBC_REENTRANT -include ../include/libc-symbol

Re: libgfortranbegin won't execute (chapter 6.16)

[Chris Staub]

On 03/31/2010 10:45 PM, Kaleb Hosie wrote:
>
> I'm at chapter 6.16 where it asked me to run the following command: "make -k 
> check"
>
> When I do that, this is the error I get after about 45minutes to an hour:
>
> No libgfortranbegin library found, will not execute fortran tests
>
>  === libgomp Summary ===
>
> # of expected passes1009
> make[4]: Leaving directory 
> `/sources/gcc-build/i686-pc-linux-gnu/libgomp/testsuite'
> make[3]: Leaving directory 
> `/sources/gcc-build/i686-pc-linux-gnu/libgomp/testsuite'
> make[3]: Entering directory `/sources/gcc-build/i686-pc-linux-gnu/libgomp'
> true  DO=all multi-do # make
> make[3]: Leaving directory `/sources/gcc-build/i686-pc-linux-gnu/libgomp'
> make[2]: Leaving directory `/sources/gcc-build/i686-pc-linux-gnu/libgomp'
> make[1]: Target `check-target' not remade because of errors.
> make[1]: Leaving directory `/sources/gcc-build'
> make: *** [do-check] Error 2
> make: Target `check' not remade because of errors.
>
>
> I know that is said in the tutorial to not stop at errors however I've 
> googled it and I couldn't find a definitive answer. Is this normal? I've 
> followed the tutorial perfectly to my knowledge.
>
> Thanks. Kaleb

You've missed the testsuite results link - 
http://www.linuxfromscratch.org/lfs/build-logs/6.6/core2duo/test-logs/076-gcc 
- in which that error does in fact appear, and it's expected since LFS 
doesn't say to build fortran anyway.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: glibc compiling problem in lfs 6.6

[Chris Staub]

On 04/13/2010 12:04 PM, Jordan Peters wrote:
> When i compile glibc, it seems to go fine but when i test it, rather
> than the output in the terminal being
> [Requesting program interpreter: /tools/lib/ld-linux.so.2]
> it's
> [Requesting program interpreter: /tools/lib64/ld-linux-x86-64.so.2]
> and this doesn't allow me to do pass 2 of binutils. why does this
> happen? i tried recompiling glibc already, and redid the specs
> assignment too, via copy/paste.
>
> thanks,
> jordan
>

I don't see the problem. If you're building x86_64, then it should be 
/tools/lib64. Also, you haven't said exactly what the problem is with 
Binutils.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: config.guess: unable to guess system type

[Chris Staub]

On 04/26/2010 10:42 AM, Jannis Kafkoulas wrote:
> Hi,
>
> I'm using LFS book 6.6 and try build lfs on a debian etch.
> I think I've done everything as described.
>
> I also downloaded the newest config scripts from git.savannah.gnu.org but the 
> message remains the same...
>
> Any idea?
>
>
> Thanks
>
There is no need to download any config.guess script. My guess is your 
host system must be missing some program that config.guess needs. Did 
you check that you have all needed packages as described on the "Host 
System Requirements" page?

Also, the Binutils source and build directories are in the wrong place - 
you should be unpacking it from $LFS/sources, not $LFS. If you've 
chowned $LFS to the lfs user or chmodded it to make it world-writable, 
change it back. Otherwise, you had to have unpacked Binutils as root, 
which you should not be doing - you must use the lfs user through all of 
Chapter 5.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: disagrees about version of symbol module_layout

[Chris Staub]

On 04/28/2010 12:42 PM, Elmar Fahrendorff wrote:
> Hi,
>
> I try to set up an LFS-System (Book Version 6.5). The System starts and
> the login messages is displayed. But at boot time the systems shows
> errors that some modules can't be loaded:
>
> Usbcore: disagrees about version of symbol module_layout
> Processor: disagrees about version of symbol module_layout
> Button: disagrees about version of symbol module_layout
> ...
>
> I used the kernel 2.6.30.2 like it is written in the LFS book.
>
>
>
> What can I do to solve this problem?
>
> Elmar Fahrendorff
>
>
> elmar.fahrendo...@miltenyibiotec.de
>

I believe this usually means that the module was compiled against a 
different kernel than the one you are trying to run. How exactly did you 
build and install the kernel and modules?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Make error in glibc

[Chris Staub]

On 05/03/2010 10:24 PM, gaurav k wrote:
> Hey Guys,
>
> This is my first mail to the list- I'm a relative newbie- though not a
> complete Linux noob who is clueless about what is happening.
>
> I have not deviated from the book at all, the only possible difference
> might be any differences in the host system requirement. For example,
> /bin/sh is not a symlink to bash, but I've only used bash throughout.
>

Yeah, that's not a problem.
>
> I went through a few of the mail archives, and I realized that I had
> made a mistake in the configparams file - I have rectified it.
>
> $ cat configparams
> CFLAGS += -march=i486 -mtune=native

The problem with configparms would definitely be the cause of the errors 
you got, so they should have gone away once you redid it. Did you remove 
the source/build directories before trying again?

>
> My extracted glibc is a subfolder of an independent folder glibc-build.

This is incorrect. Please see page 5.3, particularly in this case the 
last "Note" box. It says that for each package, you must unpack the 
package tarball, cd into the source dir, *then* follow instructions on 
the installation page. Therefore, you should already be in Glibc's 
source directory when you create the build dir.

Also, if you have managed to create /mnt/lfs/glibc-build, you had to 
have either chowned/chmodded /mnt/lfs, or created glibc-build as root 
(or using sudo), neither of which the book says to do. If you did use 
chown or chmod on /mnt/lfs, you should change it back (owned by root, 
755 perms).
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: 6.21 E2fsprogs-1.41.10 101 Tests failured -make check

[Chris Staub]

On 05/04/2010 06:28 PM, Bill Brown wrote:
> I have spend a good day trying to figure out the cause of all tests
> failing except for the first two and last two when issuing the 'make
> check' command;
>
> I am using the 6.6 version of the LFS book.  Running on Ubuntu desktop
> 9.10.  Which in turn is a virtual machine on VMware ESXi 4.0 free.  I am
> new to Linux development but have considerable console programming
> experience in C, C++, Java and earlier languages.  I am wanting to begin
> an open source development project to run on Linux so looking to broaden
> my foundation via this LFS process.
>
>
> Thanks for any insights...  Bill

That many failing tests likely indicates some kind of serious problems, 
though the output given doesn't really provide any info outside of 
saying that tests failed. There are likely clues in the log files for 
each test - check in the "tests" directory (inside the "build" dir) for 
.log files - inspect the log for each test and see if any provide useful 
information.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Make error in glibc

[Chris Staub]

On 05/05/2010 12:57 AM, gaurav k wrote:
> I'm inside the extracted glibc directory, and this is what I do:
>
> $mkdir -v ../glibc-build
> $cd ../glibc-build
> $echo "CFLAGS += -march=i486 -mtune=native">  configparams

Assuming this is a literal copy of what you actually typed, this is 
incorrect. It's "configparms" not "configparams".

> $../glibc-2.11.1/configure --prefix=/tools --host=$LFS_TGT 
> --build=$(../glibc-2.11.1/scripts/config.guess) --disable-profile 
> --enable-add-ons --enable-kernel=2.6.18 --with-headers=/tools/include/ 
> libc_cv_forced_unwind=yes libc_cv_c_cleanup=yes
> $make
>
>
>
> What I tried was, I redirected the output of the following command to a file:
>
> $../glibc-2.11.1/configure --prefix=/tools --host=$LFS_TGT 
> --build=$(../glibc-2.11.1/scripts/config.guess) --disable-profile 
> --enable-add-ons --enable-kernel=2.6.18 --with-headers=/tools/include/ 
> libc_cv_forced_unwind=yes libc_cv_c_cleanup=yes>  file.txt
>
> and so was able to see the following on the shell:
>
> configure: WARNING: using cross tools not prefixed with host triplet
> configure: WARNING:
> *** These auxiliary programs are missing or incompatible versions: msgfmt 
> makeinfo autoconf
> *** some features will be disabled.
> *** Check the INSTALL file for required versions.
> configure: WARNING: cpuid.h: present but cannot be compiled
> configure: WARNING: cpuid.h: check for missing prerequisite headers?
> configure: WARNING: cpuid.h: see the Autoconf documentation
> configure: WARNING: cpuid.h: section "Present But Cannot Be Compiled"
> configure: WARNING: cpuid.h: proceeding with the preprocessor's result
> configure: WARNING: cpuid.h: in the future, the compiler will take precedence
>
>
>
> I'm guessing THIS is the real problem?

No idea (though I think the above misspelling of configparms is the real 
issue), but the message about not finding makeinfo does also indicate 
that you've missed a page. See "Host System Requirements" and make sure 
you actually do have everything listed there.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Make error in glibc

[Chris Staub]

On 05/05/2010 07:33 AM, Simon Geard wrote:
> On Tue, 2010-05-04 at 07:54 +0530, gaurav k wrote:
>> I have not deviated from the book at all, the only possible difference
>> might be any differences in the host system requirement. For
>> example, /bin/sh  is not a symlink to bash, but I've only used bash
>> throughout.
>
> That *is* a deviation - you might be typing your commands into a bash
> shell personally, but scripts you run from there will be using /bin/sh.
> In theory programs using /bin/sh shouldn't use bash-specific features,
> but in practice, many do.
>
> Simon.
>

Then those that do should be considered buggy and must be fixed. 
Personally I think that requiring /bin/sh -> bash is pointless, and that 
if there are any packages that have /bin/sh while using Bash features 
those packages should be corrected...but then I seem to be alone on this...
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: 6.32 Inetutils-1.7: make error: telnetd link sees Undefined 'tgetent'

[Chris Staub]

On 05/21/2010 01:06 AM, Bill Brown wrote:
> LFS ver 6.6: chapter 6.32
>
> while making inetutils-1.7 I get the following error:
>
> utility.o: In function `terminaltypeok':
> /sources/inetutils-1.7/telnetd/utility.c:825: undefined reference to
> `tgetent'
> collect2: ld returned 1 exit status
> make[2]: *** [telnetd] Error 1
> 
> tgetent as I understand it is provided in the termcap package.  I tried
> installing termca-1.3.1 but that did not fix the problem.
>
> Any suggestions?
>
> Bill

You've got a problem with your Ncurses installation. What is the output 
of "ls -l /{usr/,}lib/*curses*"
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Glibc make fails with Error 2

[Chris Staub]

On 05/24/2010 12:45 AM, Kyle Brennan wrote:
> I solved the original error (it was a symlink problem with the gcc stem)
> and i got another error:
>
> mv -f /mnt/lfs/build/glibc-build/shlib.ldsT
> /mnt/lfs/build/glibc-build/shlib.lds
> i686-lfs-linux-gnu-gcc   -shared -static-libgcc -Wl,-O1  -Wl,-z,defs
> -Wl,-dynamic-linker=/tools/lib/ld-linux.so.2
> -B/mnt/lfs/build/glibc-build/csu/
> -Wl,--version-script=/mnt/lfs/build/glibc-build/libc.map
> -Wl,-soname=libc.so.6 -Wl,-z,combreloc -Wl,-z,relro
> -Wl,--hash-style=both -nostdlib -nostartfiles -e __libc_main
> -L/mnt/lfs/build/glibc-build -L/mnt/lfs/build/glibc-build/math
> -L/mnt/lfs/build/glibc-build/elf -L/mnt/lfs/build/glibc-build/dlfcn
> -L/mnt/lfs/build/glibc-build/nss -L/mnt/lfs/build/glibc-build/nis
> -L/mnt/lfs/build/glibc-build/rt -L/mnt/lfs/build/glibc-build/resolv
> -L/mnt/lfs/build/glibc-build/crypt -L/mnt/lfs/build/glibc-build/nptl
> -Wl,-rpath-link=/mnt/lfs/build/glibc-build:/mnt/lfs/build/glibc-build/math:/mnt/lfs/build/glibc-build/elf:/mnt/lfs/build/glibc-build/dlfcn:/mnt/lfs/build/glibc-build/nss:/mnt/lfs/build/glibc-build/nis:/mnt/lfs/build/glibc-build/rt:/mnt/lfs/build/glibc-build/resolv:/mnt/lfs/build/glibc-build/crypt:/mnt/lfs/build/glibc-build/nptl
> -o /mnt/lfs/build/glibc-build/libc.so -T
> /mnt/lfs/build/glibc-build/shlib.lds
> /mnt/lfs/build/glibc-build/csu/abi-note.o
> /mnt/lfs/build/glibc-build/elf/soinit.os
> /mnt/lfs/build/glibc-build/libc_pic.os
> /mnt/lfs/build/glibc-build/elf/sofini.os
> /mnt/lfs/build/glibc-build/elf/interp.os
> /mnt/lfs/build/glibc-build/elf/ld.so -lgcc
> /mnt/lfs/build/glibc-build/libc_pic.os: In function `__libc_fork':
> /mnt/lfs/build/glibc-2.11.1/posix/../nptl/sysdeps/unix/sysv/linux/i386/../fork.c:79:
> undefined reference to `__sync_bool_compare_and_swap_4'
> /mnt/lfs/build/glibc-build/libc_pic.os: In function `__nscd_drop_map_ref':
> /mnt/lfs/build/glibc-2.11.1/nscd/nscd-client.h:320: undefined reference
> to `__sync_fetch_and_add_4'
> /mnt/lfs/build/glibc-build/libc_pic.os: In function `nscd_getpw_r':
> /mnt/lfs/build/glibc-2.11.1/nscd/nscd_getpw_r.c:233: undefined reference
> to `__sync_fetch_and_add_4'
> /mnt/lfs/build/glibc-build/libc_pic.os: In function `__nscd_drop_map_ref':
>
> I looked at this output and I could not find anything aside from
> libc_pic.os seems to be throwing errors that i do not understand.
>
> Thanks,
> --Kyle

These types of errors usually means you don't have CFLAGS set correctly, 
which points to some problem with configparms. Do "cat configparms".
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: /bin/pidof error sym link in 6.53

[Chris Staub]

On 05/27/2010 06:30 AM, xinglp wrote:
> 6.53. Psmisc-22.11
>  ln -sv killall /bin/pidof
>
>
> 6.56. Sysvinit-2.86
>  sysvinit-2.86/src/Makefile:126
>  ln -sf ../sbin/killall5 $(ROOT)/bin/pidof

Not sure what you're trying to show about Sysvinit's instructions (some 
explanation would be nice) but if you're pointing out an incorrect path 
for /bin/pidof in Psmisc instructions since killall is not (by default) 
installed in /bin, that has already been corrected. Though I suppose an 
Errata entry might be good...
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: /bin/pidof error sym link in 6.53

[Chris Staub]

On 05/27/2010 01:38 PM, xinglp wrote:
> Install sysvinit-2.86 overwrite /bin/pidof.

No it doesn't. The instruction to create a /bin/pidof symlink on the 
Psmisc page is explicitly described as something to add if you *don't* 
install Sysvinit...

"If Sysvinit will not be used for a particular system, complete the 
installation of Psmisc by creating the following symlink:"
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: LFS-6.6, Stage2, glibc, nscd.c:442

[Chris Staub]

On 06/02/2010 08:05 PM, Paul Rogers wrote:
>
>
>> I don't see that the necessity of using a relatively recent version of
>> the kernel in the host environment to be, in spirit, any different
>> from requiring a relatively recent version of a C compiler. If all you
>> have is a K&R C compiler, then the solution is not to "fix" the book's
>> instructions to show how to go about building Linux with such an old
>> host. The "fix" is to upgrade the tool set.
>
> Absolutely agree, providing what you tell me is the proper toolset
> actually works.  It's not yet clear linux-2.6.18.0 works, although we
> are certainly told "Kernel too old" for 2.6.17 with the book's
> parameters.
>
> It apparently seems to some that the HSR's are an unimportant
> triviality.  Not from the user's point of view.

I wouldn't really say they are a "triviality" (given that I am one of 
those who did much of the initial testing/research for the HSR's in the 
first place) but they *are* just one of many parts of the book that 
would need to be checked, though I would argue that they are slightly 
less important than other issues (given how rarely we actually have 
issues concerning them). Package versions must work togethers, commands 
need to be correct, the book's text must be both technically correct and 
readable to those new to LFS, etc...

The HSR's are just one of many aspects of the book, and not everybody 
can test everything in complete detail. I think you simply expect too much.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: perl, gdbm and perl's obsequious help

[Chris Staub]

On 06/02/2010 05:22 PM, Neal Murphy wrote:
> Some time back (LFS 6.4), I discovered that perl's configure program can
> poison the build; it is designed to be extremely helpful by ferreting out
> features of the host system to support. Specifically in my case, because it
> found libgdbm on the host system, it configured perl to include support for
> gdbm; this causes the build to fail later because perl can't find libgdbm.
> Alas, such obsequiousness hoses the LFS build process.
>
> The standard LFS configure options for perl do not prevent this from
> happening. For review, the options are:
>  -des -Dprefix=/tools -Dstatic_ext='Data/Dumper Fcntl IO POSIX'
>
> After days of gnashing teeth and rending garments (my hair is too short to
> pull out in tufts) and parsing perl's configure script, I developed the
> correct incantation:
>  -des -Dprefix=/tools -Dstatic_ext="Data/Dumper Fcntl IO POSIX" \
>  -Donlyextensions="Data/Dumper Fcntl IO POSIX" \
>  -Dglincpth="/tools/include" -Dglibpth="/tools/lib" \
>  -Dinc_version_list=none
>
> The four extra options are required _when building the toolchain_ to ensure
> absolutely that perl's configure
>- configures those extensions and ONLY those extensions, and
>- cannot examine the host for features to support
> These options should work regardless of what is installed on the host build
> system.

I just built Perl in Chapter 5, and I do have gdbm installed (this is on 
a couple-week-old LFS svn host system). I get " NOT found." 
during Configure, and it doesn't look like anything in Perl tries to 
link to libgdbm.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: perl, gdbm and perl's obsequious help

[Chris Staub]

On 06/03/2010 11:33 AM, Neal Murphy wrote:
> On Thursday 03 June 2010 04:08:33 Simon Geard wrote:
>>
>> Curious... like Chris, I routinely build LFS from a host with gdbm
>> installed (i.e LFS itself), and never observed any problems relating to
>> perl. Indeed, even on completed LFS install, /usr/bin/perl isn't linked
>> against gdbm, and runs fine if I remove the libgdbm libraries.
>>
>> Can you be more specific about the problems you're seeing? Does the perl
>> executable fail to run at all, unable to link to libgdbm.so? Or is it
>> something less obvious?
>>
>> Simon.
>
> I built on Debian Lenny. Perl's configure examines the host system looking for
> useful features. In my case, libgdbm was installed on Lenny; perl found it
> and configured support for it. Later in the the build, perl failed to run
> because it couldn't find libgdbm.so. This was obvious. What wasn't obvious
> was, "Why?"

Still not enough info. At what point did Perl try to run, and how 
exactly? Is it just when running "perl" at all? With certain specific 
options and/or using certain Perl modules?

> If I may be so crass, that you haven't stumbled on this may be due purely to
> dumb luck. There are probably subtle differences between LFS' build steps and
> the steps I follow in my project.

And this may very well be the issue, since nobody has reported this 
problem when building LFS before. I'd say that for this report to have 
any validity, you'd need share exactly what these differences are, in 
particular how Perl is built in Chapter 5 (of course giving more than 
just "adding those 4 options to Configure"). Typically when some package 
tries to pull something from the host, it's because of a goof-up in 
Chapter 5.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Linux From Scratch (Version SVN-20100529) - 6.9. Glibc-2.11.2

[Chris Staub]

On 06/11/2010 11:35 PM, Face Man wrote:
>
> appreciated.
>
>   

First, the host system isn't an issue in Chapter 6. Second, Glibc does 
not need Autoconf. The error is complaining about ld. What is the output 
of "readelf -l /tools/bin/ld | grep interpret"
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Linux From Scratch (Version SVN-20100529) - 6.9. Glibc-2.11.2

[Chris Staub]

On 06/12/2010 12:51 AM, Face Man wrote:
>
> root:/sources/glibc-build# readelf -l /tools/bin/ld | grep interpretreadelf: 
> Error: '/tools/bin/ld': No such file
>   

What's the output of "find /tools -name 'ld*"
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Linux From Scratch (Version SVN-20100529) - 6.9. Glibc-2.11.2

[Chris Staub]

On 06/12/2010 01:04 AM, Face Man wrote:
>
> root:/sources/glibc-build# find /tools -name 'ld*'
> /tools/x86_64-unknown-linux-gnu/bin/ld
> /tools/x86_64-unknown-linux-gnu/bin/ld-old
> /tools/bin/ldd
> /tools/bin/ld-old
> /tools/x86_64-lfs-linux-gnu/bin/ld

I'm not seeing a /tools/bin/ld. Looks like you've missed a step or two, 
either in Binutils Pass 2 or in the Toolchain Adjustment (or both). It 
might be possible to mv /tools/bin/ld{-old,}, exit chroot, chown -R lfs 
$LFS/tools, and redo Binutils Pass 2, then redo the toolchain adjustment 
for ld...though I'd prefer just to go back to the beginning so you're 
more certain that it would actually work...
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Linux From Scratch (Version SVN-20100529) - 6.9. Glibc-2.11.2

[Chris Staub]

On 06/12/2010 01:55 AM, Face Man wrote:
>
> well,
> I think i should start over however i would like to know if this
> what i need to do in chapter 5.8 and 5.9
>  cd binutils-2.20.1
>
>  rm -R ../binutils-build # I am not sure i need to remove this

Um...

If this means that binutils-2.20.1 and binutils-build do in fact still 
exist, and that you have not been removing source/build directories 
after each package installation, that could very well be the source of 
your problem. Even if not, it is certainly one potential issue with your 
build and an indicator that you have missed stuff in the book and may 
have overlooked even more. Read page 5.3 *very* closely.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: LFS (Version SVN-20100529) - 5.19. File-5.04

[Chris Staub]

On 06/14/2010 06:43 AM, Face wrote:
> On Mon, Jun 14, 2010 at 1:25 PM, littlebat  wrote:
>>
>> It seems you have missed something, I remember we should always use
>> non-root user (lfs) to compile LFS in chapter 5 at least.
>>
>> See:
>> http://www.linuxfromscratch.org/lfs/view/6.6/chapter04/addinguser.html
>>
>> --
>> littlebat
>
>
> when i add lfs user i  get permissions problems

If you are getting permission errors with the lfs user, it means you've 
missed something. You need to rm -rf $LFS/tools/* and start back at the 
beginning of Chapter 4. When it says to become the lfs user, you do 
that, and you do *not* use root until it explicitly says to. If you do 
get permissions errors, you must go back to see what the problem is (or, 
of course, ask for help here), not just force root. My guess is that 
you've probably missed something else on page 5.3 - you must keep in 
mind everything mentioned in the IMPORTANT boxes on that page through 
the whole LFS book.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: How to skip two settings

[Chris Staub]

On 06/16/2010 05:34 AM, Parmenides wrote:
> Hi,
> When system starting, there are two settings, namely 'regional
> settings' and 'edit settings',
> at which the process of starting will pause and I have to press enter
> key twice to finish them.
> Is there any configurations by which I can skip them automatically every time.

An LFS system does not ask you anything (except to login) when you boot. 
This sounds like you're using the livecd, in which case you cannot 
change the startup procedure without rebuilding a new CD.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: How to skip two settings

[Chris Staub]

On 06/16/2010 01:11 PM, Parmenides wrote:
> 2010/6/16 Chris Staub:
>> On 06/16/2010 05:34 AM, Parmenides wrote:
>
>
> Yes, this is the effect of a live CD indeed. But, I have clone the
> live CD onto a partition of hard disk, and tried to altenate some
> default settings of it.

Well, the LiveCD simply isn't made for the purpose of being installed to 
a hard drive. If you want a Linux system on your hard drive you should 
just use it to build an LFS system. On the other hand, if you were 
installing the LiveCD to the hard drive in order to get around the 
time-consuming task of building LFS, you're much better off simply 
downloading and installing Ubuntu, Fedora, or some other distro.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: some issues with grub

[Chris Staub]

On 07/07/2010 01:56 PM, Bruce Dubbs wrote:
> Dr.-Ing. Edgar Alwers wrote:
>
>> 2.) the configuration file example given talks about "Be careful not
>> to change the 'exec tail' line above." . But the example does not
>> contain a exec tail line. However, this line is important.
>>
>> May be something is missing in the example ?
>
> I'll double check.
>
> -- Bruce

Nothing is missing - that is exactly what the actual file looks like. 
The comment about "exec tail" comes from /etc/grub.d/40_custom, in which 
the command does exist - grub.cfg just copies the comment lines from 
each config file and doesn't include the actual tail command.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Many certitudes and some doubt

[Chris Staub]

On 12/03/2011 04:53 AM, Giorgio Cittadini wrote:
> Hi Andy, thanks for your quick answer. I understand that the two
> solutions are equivalent: so I kept at work the symlink.
> Now, another question, yet discussed - as I could verify - in
> lfs-support, but with some residual doubt as concerns myself: the
> notorious target triplet. Running config.guess from binutils, I
> obtain, as expected, "x86_64-unknown-linux-gnu", but "echo $LFS_TGT"
> furnishes this answer: "x86_64-lfs-linux-gnu". To avoid future
> problems, must I change something (eventually, where?), or I go ahead
> not worrying about this discrepancy?
> Sorry to disturb, but I think the same question will hit other neophytes.
>

The book explicitly says to set $LFS_TGT so that "lfs" will appear in it 
- it is *supposed* to be different from the triplet given by 
config.guess - that is the point. As explained in the book, it's used to 
help with cross-compiling the initial toolchain.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] 2nd Pass binutils install fails (LFS-dev)

[Chris Staub]

On 12/05/2011 01:27 AM, str1chn1n3 wrote:
> Hello All,
> It's unclear whether the email list is restricted to just the stable
> version-- if so, just say the word and I'll abandon DEV for STABLE. If
> not, I could use a hand figuring out the cause of an install failure.
> The details:
> 1) Building to a blank system, using Debian Live CD
> (debian-live-6.0.3-i386-rescue), which includes all necessary build tools.
> 2) As stated, I'm following LFS-DEV (SVN-20111201) to the letter with a
> couple minor deviations.
> 3) Deviations to the book are limited to:
> Adding lfs user to sudoers file with 'ALL=(ALL) NOPASSWD: ALL'.

If you are following the book, this is not necessary.

> Using indirect $SPECS file creation in _5.8. Adjusting the Toolchain_ by
> doing
> $LFS_TGT-gcc -dumpspecs | sed \
> -e 's@/lib\(64\)\?/ld@/tools&@g' \
> -e "/^\*cpp:$/{n;s,$, -isystem /tools/include,}" > ./specs
> sudo mv ./specs /mnt/lfs/tools/bin/../lib/gcc/i686-lfs-linux-gnu/4.6.2

Mostly not a problem, but no need for sudo here.

> instead of the direct method
> $LFS_TGT-gcc -dumpspecs | sed \
> -e 's@/lib\(64\)\?/ld@/tools&@g' \
> -e "/^\*cpp:$/{n;s,$, -isystem /tools/include,}" > $SPECS
> used in the book.
> 4) So far all steps have executed without a hitch, until reaching the
> install step of _5.9.1. Installation of Binutils_. Then the following is
> reported:
>
> The relevant lines are:
>
> libtool: install: i686-lfs-linux-gnu-ranlib /tools/lib/libbfd.a
> ./libtool: line 1118: i686-lfs-linux-gnu-ranlib: command not found
>
> which isn't apparent to me...
> /mnt/lfs/tools lists:
>
> -rwxr-xr-x 1 root root 3397 Dec 4 21:00 catchsegv*
> -rwxr-xr-x 1 root root 65419 Dec 4 20:59 gencat*
> -rwxr-xr-x 1 root root 41023 Dec 4 20:59 getconf*
> -rwxr-xr-x 1 root root 61553 Dec 4 21:00 getent*
> -rwxr-xr-x 1 root root 2899544 Dec 4 15:53 i686-lfs-linux-gnu-addr2line*
> -rwxr-xr-x 2 root root 3022263 Dec 4 15:53 i686-lfs-linux-gnu-ar*
> -rwxr-xr-x 2 root root 4299504 Dec 4 15:53 i686-lfs-linux-gnu-as*
> -rwxr-xr-x 1 root root 2860805 Dec 4 15:53 i686-lfs-linux-gnu-c++filt*
> -rwxr-xr-x 1 root root 686890 Dec 4 20:21 i686-lfs-linux-gnu-cpp*
> -rwxr-xr-x 1 root root 60396 Dec 4 15:53 i686-lfs-linux-gnu-elfedit*
> -rwxr-xr-x 2 root root 674867 Dec 4 20:21 i686-lfs-linux-gnu-gcc*
> -rwxr-xr-x 2 root root 674867 Dec 4 20:21 i686-lfs-linux-gnu-gcc-4.6.2*
> -rwxr-xr-x 1 root root 86749 Dec 4 20:21 i686-lfs-linux-gnu-gcov*
> -rwxr-xr-x 1 root root 3300225 Dec 4 15:53 i686-lfs-linux-gnu-gprof*
> -rwxr-xr-x 4 root root 4089777 Dec 4 15:53 i686-lfs-linux-gnu-ld*
> -rwxr-xr-x 4 root root 4089777 Dec 4 15:53 i686-lfs-linux-gnu-ld.bfd*
> -rwxr-xr-x 2 root root 2920079 Dec 4 15:53 i686-lfs-linux-gnu-nm*
> -rwxr-xr-x 2 root root 3443090 Dec 4 15:53 i686-lfs-linux-gnu-objcopy*
> -rwxr-xr-x 2 root root 4036461 Dec 4 15:53 i686-lfs-linux-gnu-objdump*
> -rwxr-xr-x 2 root root 3022258 Dec 4 15:53 i686-lfs-linux-gnu-ranlib*
> -rwxr-xr-x 1 root root 627697 Dec 4 15:53 i686-lfs-linux-gnu-readelf*
> -rwxr-xr-x 1 root root 2914676 Dec 4 15:53 i686-lfs-linux-gnu-size*
> -rwxr-xr-x 1 root root 2888697 Dec 4 15:53 i686-lfs-linux-gnu-strings*
> -rwxr-xr-x 2 root root 3443089 Dec 4 15:53 i686-lfs-linux-gnu-strip*
> -rwxr-xr-x 1 root root 199282 Dec 4 20:59 iconv*
> -rwxr-xr-x 1 root root 5788 Dec 4 21:00 ldd*
> -rwxr-xr-x 1 root root 16567 Dec 4 21:00 lddlibc4*
> -rwxr-xr-x 1 root root 99174 Dec 4 20:59 locale*
> -rwxr-xr-x 1 root root 976129 Dec 4 20:59 localedef*
> -rwxr-xr-x 1 root root 6485 Dec 4 20:59 mtrace*
> -rwxr-xr-x 1 root root 23718 Dec 4 21:00 pcprofiledump*
> -rwxr-xr-x 1 root root 217377 Dec 4 21:00 rpcgen*
> -rwxr-xr-x 1 root root 4265 Dec 4 21:00 sotruss*
> -rwxr-xr-x 1 root root 62867 Dec 4 21:00 sprof*
> -rwxr-xr-x 1 root root 7133 Dec 4 20:59 tzselect*
> -rwxr-xr-x 1 root root 5374 Dec 4 21:00 xtrace*

Looks like you've been installing stuff as the root user, which I 
suspect is the reason for adding the lfs user to sudoers. This is most 
likely the cause of your problems. rm -rf /tools/*, go back to the 
beginning of Chapter 5, and do the entire chapter as the lfs user, 
without using root until the book says to.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] 2nd Pass binutils install fails (LFS-dev)

[Chris Staub]

> Whenever I attempt to install as user lfs sans sudo, I get 'Permission
> denied' for the target directory.  That's why lfs is sudoed.  Any thoughts
> why this might be happening?  (Btw, that's the reason for the indirect
> toolchain adjustment.)
>
>

That means you've missed something earlier in the book. Most likely, 
you've forgotten to chown $LFS/tools to the lfs user.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] help LFS 6.0 Ch 5.7.1 glibc-2.14.1 configure: error: gcc must provide the header

[Chris Staub]

On 12/15/2011 02:18 AM, Henk Teijema wrote:
> Guess I overlooked "The LiveCD cannot be used to build LFS 7.0 or later."
> So what liveCD should I use then?
>
> Met vriendelijke groet,
>
> Henk

Just download any fairly recent distro...Fedora, Ubuntu, etc...and if 
needed install whatever additional packages are necessary to meet the 
Host System Requirements.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Host System Versions

[Chris Staub]

On 12/25/2011 03:20 AM, Corey Richardson wrote:
> Under 'Host System Requirements', it says minimum versions. It lists
> kernel 2.6.25, but later on chapter 8 it says kernel 3.1. Is it safe to
> assume that 3.1+ is fine? Should I stick as close to those program's
> version numbers as possible or is most recent fine?

Not sure what you mean, but I guess you're confused by the Host System 
Requirements. That page only lists what's needed on the *host system*. 
The 3.1 in Chapter 8 is of course the linux kernel version you will use 
on your LFS system.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Ch6 coreutils

[Chris Staub]

On 10/11/2012 03:42 PM, Garrett Gaston wrote:
> I got this: "./bld: line 16: /usr/bin/mv: No such file or directory"
> when I ran my script for coreutils. I navigated to the /usr/bin
> directory of the chroot enironment and there is no mv file or directory.
> Any ideas? Also, if you want to see my script it's here >
> http://pastebin.com/5h9JvgHs
>

If you look closely at those commands, /usr/bin/mv is moved to /bin. The 
main problem is that you forgot to disable hashing in your scripts, so 
it's remembering the old location in /usr/bin and complaining about it 
not being there.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] bash 6.33

[Chris Staub]

On 10/11/2012 09:14 PM, Garrett Gaston wrote:
> I ran my script( http://pastebin.com/95jDZ7ds ) for chapter 6.33
> Bash-4.2 and the prompt ended up just being bash and thus never seeing
> my FLAG file or GOOD being echoed to the screen. Was my script still
> successful? I'M thinking that it was and that it just entered bash as
> part of the last line and thus didn't create my FLAG file print GOOD?
>

I think it's about time you actually learn what these commands do before 
attempting to script all this...

In this case, that line that starts with "exec" is the cause of the issue.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] glibc-2.16.0 build issue LFS-BOOK-7.2

[Chris Staub]

On 10/22/2012 02:04 PM, Gaurav Goel wrote:
> Hi,
>
> I'm following the LFS-BOOK-7.2.
> While building glibc in section 6.9 I'm getting following errors:
>
> --

> root:/sources# bash version-check.sh
> cat: /proc/version: No such file or directory

I don't know if this has anything to do with this specific problem, but 
it does indicate that something went wrong. This probably means that 
/proc isn't mounted in the chroot - does "ls /proc" give you anything?

> How can i fix this issue. Any help would be appreciated!
>
> Thanks,
> Gaurav
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] How to know how many packages are installed?

[Chris Staub]

On 10/31/2012 04:04 AM, gmspro wrote:
> Is there any way to know how many packages are installed?
> Is it possible to know if a package is installed or not?
>
> Thanks.

See the page in the LFS book on "Package Management" - 
http://www.linuxfromscratch.org/lfs/view/stable/chapter06/pkgmgt.html
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Not set up to use the correct startfiles at the 6.17 sanity check

[Chris Staub]

On 11/05/2012 09:10 PM, Harry Prevor wrote:
> This is my second LFS build, but it's been a while. This time I'm
> trying to build with Package Users as described in this hint:
> 
>
> For the sanity check at 6.17
> ,
> I get the correct program interpreter ([Requesting program
> interpreter: /lib64/ld-linux-x86-64.so.2]), but
>
>  grep -o '/usr/lib.*/crt[1in].*succeeded' dummy.log
>
> yeilds no output, which means I'm apparently not set up to use the
> correct startfiles. My dummy.log is located here:
>
>  

I see a number of suspect lines in there. In particular, a bunch that 
look like:

COMPILER_PATH=/bin/../lib/gcc/x86_64-unknown-linux-gnu/4.7.1/:/bin/../lib/gcc/
LIBRARY_PATH=/bin/../lib/gcc/x86_64-unknown-linux-gnu/4.7.1/:/bin/../lib/gcc/:/bin/../lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../lib64/:/lib/../lib64/:/usr/lib/../lib64/:/bin/../lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../:/lib/:/usr/lib/

And similarly...

attempt to open /bin/../lib/gcc/x86_64-unknown-linux-gnu/4.7.1/crtend.o 
succeeded

I see numerous similar lines in your output - basically, it appears to 
be looking in /bin and /lib for gcc files, rather than /usr/{bin,lib}. 
Does "ls /lib/gcc" give any output? Also, did you deviate from the 
book's instructions in any way other than adding Package User commands? 
Even if you did not do so deliberately, you might want to go back 
through your command history to verify what you actually typed vs. 
what's in the book.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] What Is "The" LFS Partition?

[Chris Staub]

On 11/06/2012 03:21 PM, Feuerbacher, Alan wrote:
> Bruce wrote:
>
>>> Why does one have to create a directory with that name before
>>> executing the "mount" command?
>>
>> The system has to know where to attach the data structures in the file
>> tree.  You could create a script to do a 'mkdir -p ;
>> mount...', but that's overkill.
>
> Now I'm confused again. I thought that creating a directory actually writes 
> data into a place on a hard disk that the kernel allocates for the directory. 
> Something about inodes, if I remember right. But if that's so, and a 
> filesystem is not yet mounted, where does that data get written? It looks 
> like the cart is before the horse.
>
> Specifically, if you want to do "mount /dev/sda5 /mnt/lfs", but you have to 
> create the directory "/mnt/lfs" BEFORE you do the mount, then where does the 
> inode information about "/mnt/lfs" get written? I'm sure I'm missing some 
> details.

I don't see why there is any confusion here. Before you mount to a 
directory, the mount point is...just a directory. I don't understand the 
question about directory inodes with mounting - a created directory does 
of course take up some space on the disk, same as any other file. I 
think you are just seeing additional complications where there is none.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Not set up to use the correct startfiles at the 6.17 sanity check

[Chris Staub]

On 11/06/2012 08:13 PM, Harry Prevor wrote:
>
> On 11/5/12, Chris Staub  wrote:
>> I see numerous similar lines in your output - basically, it appears to
>> be looking in /bin and /lib for gcc files, rather than /usr/{bin,lib}.
>
> To be honest, I'm not really sure why that is.
>
>> Does "ls /lib/gcc" give any output?
>
> After recompiling as root it does (not sure whether it did before but
> I believe it would have): "x86_64-unknown-linux-gnu"

That directory should not exist. This would seem to indicate some 
mistyped directory option, but your dummy.log contents say that GCC was 
configured with --prefix=/usr and --libexecdir=/usr/lib, which are both 
correct. Installing GCC into /usr certainly shouldn't create any 
directories or files in /bin or /lib. Do you have logs of your GCC 
installation?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Not set up to use the correct startfiles at the 6.17 sanity check

[Chris Staub]

On 11/06/2012 09:11 PM, Harry Prevor wrote:
> On 11/6/12, Chris Staub  wrote:
>
>> Installing GCC into /usr certainly shouldn't create any
>> directories or files in /bin or /lib. Do you have logs of your GCC
>> installation?
>
> Only of the installation as root, as all the previous files were
> deleted. Here is my config.log:
>
>  <http://sprunge.us/QUHB>
>
> I can provide any other files from the installation if requested.
>
configure:3785: found /bin/gcc

Looks like you didn't remove the previous GCC installation. Either that, 
or that was there even before you tried building GCC in chapter 6, which 
would point to some issue while making symlinks/files early in Chapter 
6. If /bin/gcc is owned by the gcc user (most likely), rm it, and 
everything else owned by that user (perhaps with the uninstall_package 
script). Then try the sanity check again.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Not set up to use the correct startfiles at the 6.17 sanity check

[Chris Staub]

On 11/06/2012 09:19 PM, Chris Staub wrote:
> On 11/06/2012 09:11 PM, Harry Prevor wrote:
>>
>> Only of the installation as root, as all the previous files were
>> deleted. Here is my config.log:
>>
>>   <http://sprunge.us/QUHB>
>>
>> I can provide any other files from the installation if requested.
>>
> configure:3785: found /bin/gcc
>
> Looks like you didn't remove the previous GCC installation. Either that,
> or that was there even before you tried building GCC in chapter 6, which
> would point to some issue while making symlinks/files early in Chapter
> 6. If /bin/gcc is owned by the gcc user (most likely), rm it, and
> everything else owned by that user (perhaps with the uninstall_package
> script). Then try the sanity check again.

I just took another look, and saw this...

configure:8047: found /bin/ar
configure:8058: result: ar
configure:8172: checking for as
configure:8188: found /bin/as
configure:8199: result: as
configure:8313: checking for dlltool
configure:8343: result: no
configure:8454: checking for ld
configure:8470: found /bin/ld
configure:8481: result: ld
configure:8595: checking for lipo
configure:8625: result: no
configure:8736: checking for nm
configure:8752: found /bin/nm
configure:8763: result: nm
configure:8877: checking for ranlib
configure:8893: found /bin/ranlib
configure:8904: result: ranlib
configure:9013: checking for strip
configure:9029: found /bin/strip
configure:9040: result: strip

Apparently Binutils was also installed into /bin. I'd rm all of Binutils 
and GCC and reinstall both, and of course double-check command history 
for Binutils as well.

Out of curiosity, are lib{gmp,mpfr,mpc} also in /lib, or /usr/lib?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] libmpfr error while compiling GCC.

[Chris Staub]

On 11/09/2012 12:30 PM, Prabhu wrote:
>
> Hi,
>   I'm trying to compile GCC, I done with untaring (MPFR,GMP and
> MPC ) and moving the files to its respected folders, and even
> configuration part also done. but while making the Gcc its throwing me
> an error message says that

Please clarify what exactly you did. Specifically, what is meant by 
"moving the files to its respected folders" as LFS never says to move 
anything for MPFR - only to unpack and rename the directory.

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] libmpfr error while compiling GCC.

[Chris Staub]

On 11/10/2012 01:49 AM, Prabhu wrote:
> Hi,
>   I'm trying to compile GCC, I done with untaring (MPFR,GMP and
> MPC ) and moving the files to its respected folders, and even
> configuration part also done. but while making the Gcc its throwing me
> an error message says that
>
> *checking for MPFR... no*
> *configure: error: libmpfr not found or uses a different ABI (including
> static vs shared).*
> *make[1]: *** [configure-mpc] Error 1*
> *make[1]: Leaving directory `/mnt/test/sources/gcc-build'*
> *make: *** [all] Error 2*
>
>
> please let me know where I failed. I tried almost 4 times from the
> starting, its keep on failing me in same stage.
>
>
>
> With Regards...
>  PRABHU :)

Please specify what you mean. Give every command you typed, in order, 
for the GCC installation.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] libmpfr error while compiling GCC.

[Chris Staub]

On 11/10/2012 07:14 AM, Prabhu wrote:
> (5.5.1 Installation of cross GCC)
>
> I successfully compiled Binutilities, then I untared the GCC. In GCC
> folder I done the following steps
>
> *tar -Jxf ../mpfr-3.1.1.tar.xz
> mv -v mpfr-3.1.1 mpfr
> tar -Jxf ../gmp-5.0.5.tar.xz
> mv -v gmp-5.0.5 gmp
> tar -zxf ../mpc-1.0.tar.gz
> mv -v mpc-1.0 mpc*
>
> Now the final step is to compile GCC by *make *command, at this point of
> time it got compiled for 10 mins and then it was showing an error
> message that *
>
> *checking for MPFR... no*
> *configure: error: libmpfr not found or uses a different ABI (including
> static vs shared).*
> *make[1]: *** [configure-mpc] Error 1*
> *make[1]: Leaving directory `/mnt/test/sources/gcc-build'*
> *make: *** [all] Error 2*
>
> please let me know where I failed. I tried almost 4 times from the
> starting, its keep on failing me in same stage.
>
> With Regards...
>  PRABHU :)

Everything you did looks correct, or at least done in the right order - 
you did unpack mpfr in the gcc source dir, so that should work fine. 
However, everything you pasted seems as if you just copied from the book 
into the email - perhaps there's something you're doing when actually 
inputting the command. Can you verify by checking your exact commands 
from your command history? Pasting the contents of 
gcc-build/mpc/config.log may also help.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] LFS 7.2 Section 6.4 Chroot Error

[Chris Staub]

On 11/10/2012 10:05 AM, yan wrote:
> I also get a problem at this point
>
>chroot "$LFS" /tools/bin/env -i HOME=/root TERM="$TERM" PS1='\u:\w\$ '
> PATH=/bin:/usr/bin:/sbin:/usr/sbin:/tools/bin /tools/bin/bash --login +h
> /tools/bin/env: /tools/bin/bash: Permission denied
>
>
> root@cpu:/home/yan/lfs# ll /tools/bin/env
> -rwxrwxr-x 1 lfs lfs 102693 Oct 28 12:24 /tools/bin/env*
>
> root@cpu:/home/yan/lfs# readelf -e /tools/bin/env | grep interpreter
> [Requesting program interpreter: /tools/lib/ld-linux.so.2]
>
> I don't have any idea to solve it

Type "ls -la /tools/bin/bash"
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] LFS 7.2 Section 6.4 Chroot Error

[Chris Staub]

On 11/10/2012 09:22 PM, yan wrote:
> On Sat 10 Nov 2012 11:09:08 PM CST, Chris Staub wrote:
>> On 11/10/2012 10:05 AM, yan wrote:
>>> I also get a problem at this point
>>>
>>>  chroot "$LFS" /tools/bin/env -i HOME=/root TERM="$TERM" PS1='\u:\w\$ '
>>> PATH=/bin:/usr/bin:/sbin:/usr/sbin:/tools/bin /tools/bin/bash --login +h
>>> /tools/bin/env: /tools/bin/bash: Permission denied
>>>
>>>
>>> root@cpu:/home/yan/lfs# ll /tools/bin/env
>>> -rwxrwxr-x 1 lfs lfs 102693 Oct 28 12:24 /tools/bin/env*
>>>
>>> root@cpu:/home/yan/lfs# readelf -e /tools/bin/env | grep interpreter
>>>   [Requesting program interpreter: /tools/lib/ld-linux.so.2]
>>>
>>> I don't have any idea to solve it
>>
>> Type "ls -la /tools/bin/bash"
> thank you for the prompt
> it's ok when i add +x permission for all file in directory /tools/bin/
>
It is not ok, since that should never be necessary. You could continue 
on if you want to, but note that since you still don't know why it 
actually failed, there's no telling if you will have more problems later on.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] GCC-4.7.1 - Pass 2 Error unrecognized command line option '-V' unrecognized command line option '-qversion'

[Chris Staub]

On 11/12/2012 10:25 AM, lei huang wrote:
> Linux  2.6.18-308.16.1.el5 #1 SMP Tue Oct 2 22:01:37 EDT 2012 i686
> i686 i386 GNU/Linux
>
> config.log
>
> FATAL: kernel too old
>
> help!!!

There's the problem - your host system doesn't have a recent enough 
kernel. You will need to upgrade the kernel to at least the minimum 
version specified on the "Host System Requirements" page.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Use a newer kernel than specified?

[Chris Staub]

On 11/14/2012 04:59 AM, Smith, Mitchell wrote:
> Hi,
>
> The package list specifies using linux kernel: Linux-3.5.2
>
> Obviously the later releases would contain multiple fixes and security
> improvements.
>
> Any reason why using the latest 3.5.X release (Currently 3.5.7) would
> cause any issues
>
> Thanks
>
> --
> *Mitchell

 From the package download page in the book:

"The Linux kernel is updated relatively often, many times due to 
discoveries of security vulnerabilities. The latest available 3.5.x 
kernel version should be used, unless the errata page says otherwise."
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] LFS network configuration for internet access

[Chris Staub]

On 11/16/2012 01:07 AM, Oshadha Gunawardena wrote:
> Hi all,
>
> Since I have completed building complete LFS system now I want to
> configure it's network in order to access internet.
>
> I have read through
>
>  take new kernel, copy old config to .config
>  make oldconfig
>  answer the questions it asks
>  if the kernel versions are similar, there will be fairly few
> questions
>  after that, if you were NOT presented with the choice you wanted
>  go and turn it on
>
> ls /sys/class/net
>
> cd to the kernel source dir and try again
> look at the output carefully

What's your question?

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] LFS-7.2 host system tar version

[Chris Staub]

On 11/23/2012 07:33 PM, nettxzl wrote:
> (Sorry if this comes through twice. My first attempt seems to have vanished.)
>
> Hi,
>
> I was starting the gcc-4.7.1 (Pass 1) build in Chap. 5 and I was
> trying to unpack the files compressed with xz following the command
> given there:
>
> tar -Jxf ../mpfr-3.1.1.tar.xz
>
> but I got this error message:
>
> tar: invalid option -- J
>
> I guessed my host system tar is too old, but it is tar-1.19 and this
> is newer than the minimum version, tar-1.18, stated in the host system
> requirements.
> It's not a problem because I can unpack the files in two steps: first
> with xz, then with tar.
>
> But I think you may want to update the tar version in the host system
> requirements.

I just confirmed this with a built-from-source vanilla tar 1.20, which 
is also newer than the current Host System Requirments state - indeed it 
does not support xz archives, and the book should certainly be updated 
to specify a newer minimum version.

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] "-z option ...no" issue

[Chris Staub]

On 11/26/2012 12:19 AM, Bruce Dubbs wrote:
> 王杰清 wrote:
>> hi,
>> when i install glibc following lfs.7.2 book, i meet with the error
>> "configure: error : linker with -z support required"
>> is my gcc  wrong in my host? should i reinstall it ,or how can i fix it .
>> Any advice is pleasure;
>
> Have you checked the host system requirements?
>
> -- Bruce
>
Host System, at least in regards to compiler/linker, should be 
irrelevant by the time you're at Glibc. What's the output of "env" as 
the lfs user? Also, "ls -l /tools/bin".
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] next step wireless

[Chris Staub]

On 11/27/2012 05:15 PM, Dave wrote:
> Dr.-Ing. Edgar Alwers wrote:
>> On Tuesday 27 November 2012 11:09:33 Dave wrote:
>>
>>> I only have  WiFi with WEP encryption here, can the WICD pkg and DHCP
>>> alone handle this connect?
>>>
>>
>> You will need some other packages, like wpa_supplicant, but the answer is 
>> yes.
>> It is late here, and I am tired, but I can come back to the issue if you need
>> advice. I am working with WICD  and DHCP.

If you're just using WEP, then you don't really need anything other than 
wireless_tools. Of course you also want dhcpcd if you're connecting via 
dhcp, but that's actually separate from the wireless settings. Just 
install wireless_tools (see BLFS) and type "iwconfig [wlan0] essid 
[SSIDof_your_router] enc [encryption_key]" before bringing up your 
network connection - and maybe add that to your network startup script.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Help with a linker problem

[Chris Staub]

On 11/30/2012 06:37 AM, Simon Geard wrote:
> Hey guys...
>
> Rewriting my LFS build scripts, I'm getting an interesting linking error
> with bash in chapter 6, and am hoping someone can point me in the right
> direction for tracking down the cause.
>
> Basically, bash builds correctly, but the resulting binary is bad, any
> attempt to invoke it resulting in:
>
> # /bin/bash --version
> /bin/bash: error while loading shared libraries: libncurses.so.5: cannot
> open shared object file: No such file or directory
>
> Now, the obvious problem is that it's complaining about libncurses, not
> the libncursesw version built in chapter 6. As far as I can tell, it
> should be linked to the latter, thanks to the "INPUT(-lncursesw)" linker
> script created when we installed ncurses.
>
> My suspicion is that when the bash binary was linked, the linker found
> ncurses in /tools ahead of /lib, and so not applied that linker script.
> It then fails runtime linking, since /tools isn't in the search path at
> runtime (unless I put it there with LD_LIBRARY_PATH).
>
>
> So, it seems pretty clear that I've missed something in my script -
> probably in one of those first few parts of chapter 6. But I can't spot
> the problem myself, so I'm hoping someone can point me in the direction
> of what could be causing this.
>
> Simon.

First, check how Bash is linked: "readelf -l /bin/bash | grep 
interpret". Of course, this should say that it's looking for the dynamic 
linker in /lib. Then verify you actually have all the right libraries 
for Ncurses: "ls -la {/usr,}/lib/*ncurses*"

You could also just upload your build scripts for Ncurses, Readline, and 
Bash.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] /dev/pts and /dev/shm not mounted on boot

[Chris Staub]

On 12/02/2012 01:02 AM, d...@fluidnetwork.com wrote:
> Hello all,
>
> Just built two systems, one of HLFS-development and one LFS-7.2.
>
> Not sure what I did wrong, but neither system would successfully mount
> /dev/pts or /dev/shm on bootup - both were complaining about
> nonexistant mount points.  After searching the web for a few hours, I
> came across one lonely half suggestion for a fix... and it works.
>
> In the file /etc/rc.d/init.d/mountfs
> I added the following at line 38
>  (   belowboot_mesg "Mounting remaining file systems"  )
>  mkdir /dev/{pts,shm}
>
> It worked.
>
> DJ

That should not be needed. I believe this points to a problem with Udev, 
as it should create /dev/pts when it starts. Did you deviate from the 
book's instructions in any way?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] error when i build webkitgtk-1.10.1

[Chris Staub]

On 12/02/2012 11:22 AM, Francisco Jesús Navarro Cortés wrote:
> Hello.  i try build with make-3.82 and make-3.82-upstream_fixes-3.patch
> the source webkitgtk-1.10.1 but get this error:
>
> ./.libs/libwebkit2gtk-3.0.so  : undefined 
> reference to `glCullFace'
> ./.libs/libwebkit2gtk-3.0.so  : undefined 
> reference to `glGetTexParameterfv'
> collect2: error: ld returned 1 exit status
> make[1]: *** [Programs/WebKitWebProcess] Error 1
> make[1]: se sale del directorio `/sources/webkitgtk-1.10.1'
> make: *** [all] Error 2
>
> I sorry for my bad english.
>
> Thanks.

Looks like it's not finding GL stuff from Mesa. What do you get if you 
do "ls -l /usr/include/GL"?

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Permission error in Binutil while "make install" (LFS7.2)

[Chris Staub]

On 12/08/2012 03:49 PM, Prabhu wrote:
> Hi I'm getting permission error while "make install" the binutil
> package, but I could successfully able to create folder in all the 3 dir
> (source,tools,usr) with the user lfs. I rollback and reperformed the
> chapter 4.1 to 4.4, but I'm still facing the issue.
>
>
> *lfs@logun-HP-Pavilion-dv6-Notebook-PC:/mnt/lfs/sources/binutils-build$
> make install
> make[1]: Entering directory `/mnt/lfs/sources/binutils-build'
> /bin/sh ../binutils-2.22/mkinstalldirs
> /tools--with-sysroot=/mnt/lfs--with-lib-path=/tools/lib--target=i686-lfs-linux-gnu
> /tools--with-sysroot=/mnt/lfs--with-lib-path=/tools/lib--target=i686-lfs-linux-gnu
> mkdir -p --
> /tools--with-sysroot=/mnt/lfs--with-lib-path=/tools/lib--target=i686-lfs-linux-gnu
> /tools--with-sysroot=/mnt/lfs--with-lib-path=/tools/lib--target=i686-lfs-linux-gnu
> mkdir: cannot create directory `/tools--with-sysroot=': Permission denied
> mkdir: cannot create directory `/tools--with-sysroot=': Permission denied

That is the problem - it is trying to create a "/tools--with-sysroot=" 
directory. Apparently you've mistyped the configure command and left out 
spaces. Just try again making sure to enter the commands correctly.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Installation of Bash 5.15 Error on make

[Chris Staub]

On 12/15/2012 02:58 AM, Alexander Spitzer wrote:
> Hello all,
>
> I am on section 5.15.1 (installing bash) on LFS version 7.2 and I am
> receiving an error after running make:
> $ make -j
> yacc -d ./parse.y
> make: execvp: yacc: Permission denied
> make: *** [y.tab.c] Error 127
>
> I have encountered this before and I attempted to get around by sudo,
> but that turned out to be a bad idea as the bash binaries were linked to
> libraries on the host system. After doing a fresh start I am once again
> confronted by this error.
>
> I suspect this has something to do with the yacc link.
> readlink -f /usr/bin/yacc -> /usr/bin/yacc
>
> And here is /usr/bin/yacc:
> #! /bin/sh
> exec '/usr/bin/bison' -y "$@"
>
> Is this not a script executing bison as specified in the host
> requirements? If not, how should /usr/bin/yacc look?
>
> Is there anything else that may be causing the permission denied error?
>
> Thanks,
> Alex

I doubt that's a problem at all. Looks like it's having issues running 
the script at all, which wouldn't have anything to do with Bison itself. 
Do "ls -l /usr/bin/yacc"
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Installation of Bash 5.15 Error on make

[Chris Staub]

On 12/15/2012 03:03 AM, Alexander Spitzer wrote:
> An additional piece of information that might be important:
> The patch asked for in the book is bash-4.2-fixes-8.patch while the one
> I have from the website and that one that I applied
> is bash-4.2-fixes-10.patch
> Could that be the cause of the problem?

That is extremely unlikely, though since you are saying you are doing 
something differently than what's in the book, I need to ask how else 
(no matter how trivial it might seem) you are deviating from the book's 
instructions. Also, you're saying the book mentions a "fixes-8" patch as 
stated in the 7.2 book - that specific patch is in fact available - it's 
even linked right in the book.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Installation of Bash 5.15 Error on make

[Chris Staub]

On 12/15/2012 03:11 AM, Alexander Spitzer wrote:
> Hi Chris,
> Thanks for your reply.
>
> ls -l /usr/bin/yacc gives:
> -rw-r--r-- 1 root root 41 Dec 13 19:13 /usr/bin/yacc
>
>
That explains it, that script is not executable. "chmod a+x 
/usr/bin/yacc" and it should work.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Installation of Bash 5.15 Error on make

[Chris Staub]

On 12/15/2012 03:18 AM, Alexander Spitzer wrote:
> I used the sources in wget-list and the list had the fixes-10.patch. I
> hope that doesn't turn out to be the issue. Other than that, the only
> thing I can think of is during the make of the check package, I received
> a undefined reference to pthread. error in the directory tests,
> which I fixed by:
> cd tests
> open the makefile and add -lpthread to the variable "CFLAGS"
> after that it built properly.
>
> The book mentions the swap partition but I only made 1 ext3 3GB sized
> partition on a 4GB USB stick.

The wget-list in LFS 7.2 does in fact also have the "fixes-8" patch, so 
it looks like you are doing something weird with mixing 7.2 and the 
development book. I don't know if this is the specific cause of any of 
your issues, but first thing I'd suggest would be to go back to the 
beginning and only use 7.2.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Binutils patch, why?

[Chris Staub]

On 12/16/2012 06:55 AM, JIA Pei wrote:
>
> Hi, all:
>
> 1) From
> http://www.linuxfromscratch.org/lfs/view/stable/chapter05/binutils-pass1.html
> , binutils pass 1 requires a patch, why?
> What is this patch for?
>
> *patch -Np1 -i ../binutils-2.22-build_fix-1.patch*
>
It tells you in the book. Try reading it.

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Is it a must to separate Binutils and GCC into Pass 1 and Pass 2?

[Chris Staub]

On 12/17/2012 06:00 AM, JIA Pei wrote:
>
> Hi, all;
>
> Sorry to bug all of you again.
> I've successfully built
> Binutils  2.23.1
> GCC   4.7,2
> Linux-kernel   3.6.10
> Glib   2.16.0
>
> However, is it a must for me to proceed to the pass 2 of binutils and gcc?

Whatever is in the book is there for a reason.
>
> I'm now following "Pass 2 of Binutils "
> http://www.linuxfromscratch.org/lfs/view/stable/chapter05/binutils-pass2.html
>
> the following step
> /CC=$LFS_TGT-gcc\/
> /AR=$LFS_TGT-ar \/
> /RANLIB=$LFS_TGT-ranlib \/
> /../binutils-2.22/configure \/
> /--prefix=/tools\/
> /--disable-nls  \/
> /--with-lib-path=/tools/lib/
>
> brings me error messages:
>
> /lfs@peijia-GA-870A-UD3:/mnt/lfs/sources/binutils-build$ CC=$LFS_TGT-gcc
> \/
> /> AR=$LFS_TGT-ar \/
> /> RANLIB=$LFS_TGT-ranlib \/
> /> ../binutils-2.23.1/configure \/
> /> --prefix=/tools\/
> /> --disable-nls  \/
> /> --with-lib-path=/tools/lib/
> /checking build system type... i686-pc-linux-gnu/
> /checking host system type... i686-pc-linux-gnu/
> /checking target system type... i686-pc-linux-gnu/
> /checking for a BSD-compatible install... /usr/bin/install -c/
> /checking whether ln works... yes/
> /checking whether ln -s works... yes/
> /checking for a sed that does not truncate output... /bin/sed/
> /checking for gawk... gawk/
> /checking for gcc... i686-lfs-linux-gnu-gcc/
> /checking for C compiler default output file name... a.out/
> /*checking whether the C compiler works... configure: error: in
> `/mnt/lfs/sources/binutils-build':*/
> /*configure: error: cannot run C compiled programs.*/
> /If you meant to cross compile, use `--host'./
> /See `config.log' for more details./
>
>
> So, my question now is: is Pass 2 of binutils a must??

You are clearly having great difficulty. For one thing, you claim you 
are following the stable book but you are using versions from the latest 
development, and obviously that is not working well, and of course that 
doesn't take into account whatever other deviations from the book that 
you might be doing that you haven't mentioned. Solution: rm -rf 
$LFS/tools, go back to page 1, actually *read* every word in each page 
of the book, not just the commands (as most of your questions so far 
have in fact been answered right in the book itself), and do exactly as 
instructed, including the correct package versions.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Is it a must to separate Binutils and GCC into Pass 1 and Pass 2?

[Chris Staub]

On 12/18/2012 07:51 AM, JIA Pei wrote:
>
> Hi, Simon:
>
> The reason why I'm doing that (use binutils-2.23.1 instead of
> binutils-2.22) is I don't want to 100% strictly follow LFS book, so that
> I might be able to understand how to build my own Linux deeper.
>
> And, even now, I've got no idea why we need to patch binutils ??
> Because without patching binutils, I'm still able to pass BinUtils Pass
> 1

The book itself quite clearly explains what the patch does. If you 
*really* want to "learn" then start by, for example, reading the patch 
itself to see how exactly it works.

However, since you still felt to need to ask, for example, if 
Binutils/GCC Pass 2 was necessary, despite the fact that the build 
process is thoroughly explained, it appears you aren't really making 
much of an effort to read the information that's already there. If you 
really do want to "learn" start off by - as I've said to you at least 
twice before - actually reading all the text in the book.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Glibc installation in target machine fails

[Chris Staub]

On 12/18/2012 08:25 AM, Prabhu wrote:
> Hi,
>
> Since its a check, I ignored it and I performed with the next step
> that's "*make install*" even this fails and throw me an error as
> *
> *ERROR*:
>
> make[2]: *** No rule to make target
> `/usr/--disable-profile--enable-add-ons--enable-kernel=2.6.25/lib/libBrokenLocale.so',
> needed by `install-lib-nosubdir'.  Stop.
> make[2]: Leaving directory `/sources/glibc-2.16.0/locale'
> make[1]: *** [locale/subdir_install] Error 2
> make[1]: Leaving directory `/sources/glibc-2.16.0'
> make: *** [install] Error 2*
>
> could someone assist me resolve this.
> --
>
> With Regards...
>  PRABHU :)

Looks like it thinks it was given an odd prefix. There was probably a 
typo on the configure command. Can you check your command history and 
paste the exact configure command you gave?

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Is it a must to separate Binutils and GCC into Pass 1 and Pass 2?

[Chris Staub]

On 12/18/2012 09:23 PM, JIA Pei wrote:
>
> Now, fail to build check-0.9.9 ...
> The error message is:
>
> //mnt/lfs/tools/bin/../lib/gcc/i686-pc-linux-gnu/4.7.2/../../../../i686-pc-linux-gnu/bin/ld:
> note: 'pthread_create@@GLIBC_2.1' is defined in DSO
> /tools/lib/libpthread.so.0 so try adding it to the linker command line/
> //tools/lib/libpthread.so.0: could not read symbols: Invalid operation/
> /collect2: error: ld returned 1 exit status/
> /make[2]: *** [check_thread_stress] Error 1/
> /make[2]: Leaving directory `/mnt/lfs/sources/check-0.9.9/tests'/
> /make[1]: *** [all-recursive] Error 1/
> /make[1]: Leaving directory `/mnt/lfs/sources/check-0.9.9'/
> /make: *** [all] Error 2/
>
>
> I tried to re-configure it as mentioned in
> http://comments.gmane.org/gmane.linux.lfs.support/35698
> namely:
> *$ CFLAGS="-L/tools/lib -lpthread"*
> *$ ./configure --prefix=/tools*
>
>
> Problem continues, namely, the above error message is still there...
>
> How to solve this problem though?

You want to "learn", you figure out yourself how to fix it.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Is it a must to separate Binutils and GCC into Pass 1 and Pass 2?

[Chris Staub]

On 12/18/2012 09:07 PM, JIA Pei wrote:
>
> Thank you so much Bruce...
> I proceed to expect 5.45 now 
> But, I got two test fails as follows. Is this normal?
>

If you had read the book like you're expected to, you would already know 
the answer to this. If you can't be bothered to read it, quit wasting 
time here and figure it all out yourself.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] make install gcc 4.7.2 fail...

[Chris Staub]

On 12/20/2012 06:33 PM, JIA Pei wrote:
>
> However, I *failed to make install gcc 4.7.2*
>
> The error message obtained is:
>
> /* tar -cf - .; exit 0) | (cd
> /usr/lib/gcc/i686-pc-linux-gnu/4.7.2/include; tar xpf - )
> /bin/sh: line 1: tar: command not found
> /bin/sh: line 1: tar: command not found
> make[2]: *** [install-headers-tar] Error 127
> make[2]: Leaving directory `/sources/gcc-build/gcc'
> make[1]: *** [install-gcc] Error 2
> make[1]: Leaving directory `/sources/gcc-build'
> make: *** [install] Error 2*/
>
>
> This is understandable for me because after *chroot* , I don't have the
> command tar under /usr/bin any longer. And according to the installation
> sequence listed at
> http://www.linuxfromscratch.org/lfs/view/development/index.html,
> *Tar-1.26* is far way behind *GCC-4.7.2* .

I can't help but notice the lack of surprise at tar not being found in 
/tools. Did you happen to skip it? Also, if tar didn't work in chroot, 
how have you been able to unpack the tarballs for GCC and every other 
package in Chapter 6?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Is it normal that I obtained a lot of error messages while building glibc-2.16.0?

[Chris Staub]

On 12/22/2012 06:30 AM, JIA Pei wrote:
>
> Hi, all:
>
> I noticed on
> http://www.linuxfromscratch.org/lfs/view/development/chapter06/glibc.html ,
> I may meet some building errors while building glibc. I obtained the
> following errors. I just wonder if as mentioned on the above page, can I
> just ignore all the following error messages??
>
>
> Thanks

First explain exactly what you are doing. Your previous question implied 
that you did not install tar into /tools, for starters. Is that 
accurate, and what else are you doing differently?

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] make install gcc 4.7.2 fail...

[Chris Staub]

On 12/22/2012 03:37 PM, JIA Pei wrote:
>
> Hi, Christ:
>
>
> I recalled it now... This problem is probably caused by the SD
> card automatic disconnection from the computer during *Installing Basic
> System Software --- probably happened after "make" at page
> *http://www.linuxfromscratch.org/lfs/view/development/chapter06/gcc.html. In
> such a case, I've got no idea*what are the required steps I need to
> re-run before re-mount the SD Card*.
> Apparently, it's not just simple remount the SD card by a command line,
> it has something to do with chroot, and environment re-configuration I
> guess --- Do I require to re-run
> http://www.linuxfromscratch.org/lfs/view/development/chapter06/kernfs.html?
>
> Looking forwarding to your reply

The book tells you.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] LFS 7.2 section 6.52 kmod-9

[Chris Staub]

On 12/26/2012 02:44 AM, Israel Silberg wrote:
> Hi all,
> I'm building LFS 7.2, and during the running of the configure section of
> the kmod-9 I got the following:
>
> checking for liblzma... no
> configure: error: Package requirements (liblzma >= 4.99) were not met:
>
> No package 'liblzma' found
>
> Consider adjusting the PKG_CONFIG_PATH environment variable if you
> installed software in a non-standard prefix.
>
> Alternatively, you may set the environment variables liblzma_CFLAGS
> and liblzma_LIBS to avoid the need to call pkg-config.
> See the pkg-config man page for more details.
>
> How can I fix it? during the compilation of xz I didn't get any error
> messages.
> Thanks,
> Israel

Paste the contents of config.log.

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] LFS 7.2 section 6.52 kmod-9

[Chris Staub]

> You should have liblzma.so in /tools/lib and lzma.h in /tools/include.
>
>-- Bruce
> --
> http://linuxfromscratch.org/mailman/listinfo/lfs-support
> FAQ: http://www.linuxfromscratch.org/lfs/faq.html
> Unsubscribe: See the above information page
>
>
> Hi Bruce,
> Yes bith files are in their described place but they are not under
> /usr/lib (shouldn't they be since xz is already complied in chapter 6?
>
> Israel

liblzma should be in /lib, not /usr/lib. Also, can you copy the contents 
of config.log?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] LFS 7.2 section 6.52 kmod-9

[Chris Staub]

On 12/26/2012 05:25 PM, Baho Utot wrote:
> On 12/26/2012 02:10 PM, Chris Staub wrote:
>> liblzma should be in /lib, not /usr/lib. Also, can you copy the contents
>> of config.log?
>
>   From 6.46. Xz-5.0.4
>
> ./configure --prefix=/usr --libdir=/lib --docdir=/usr/share/doc/xz-5.0.4
>
> the prefix puts it into /usr/lib
>
> usr/lib/
> usr/lib/liblzma.a
> usr/lib/liblzma.so
> usr/lib/liblzma.so.5
> usr/lib/liblzma.so.5.0.4
> usr/lib/pkgconfig/
> usr/lib/pkgconfig/liblzma.pc
>
>
Note the "--libdir=/lib".
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] How to upgrade a single package?

[Chris Staub]

On 01/03/2013 04:13 AM, JIA Pei wrote:
>
> Hi, all:
>
> sorry for my naive question again.
> I now noticed coreutils has been upgraded from 8.19 to 8.20 . I'd love
> to upgrade it without rebuilding everything from scratch, but just this
> single package.
>
> I can always chroot into the environment required by chapter 6, but how
> can I re-enter chapter 5 to prepare the tools coreutils using version
> 8.20 rather than version 8.19 ?

You can't even get a basic LFS system built right, you don't have any 
hope of successfully upgrading anything. Further, you've asked one 
question ("how to upgrade a package") that's answered in the FAQ, and 
another ("how to re-enter chroot") that's answered in the book itself.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] can't compile gcc pass 1

[Chris Staub]

On 01/05/2013 07:35 AM, Fernando de Oliveira wrote:
>
> I have recurrently written about this, here.
>
> My error with this page, even after having built one or two versions of
> LFS, was that the last line:
>
> "gcc compilation OK"
>
> that made me ignore the other lines, when some of these lines were
> telling me that I had requirements to fix.
>
> I thought: "all right, host can compile, so the other lines are just
> additional information or recommendations, may be".
>
> My suggestion in a previous post was to move that line to an earlier
> place in the script.
>
> Chris gave better suggestions than mine. But as the SVN version has
> been modified, Bruce does not think that anything is necessary, "I have
> modified the book, let us wait until new release on March". Having read
> that before my original post, I told him that we could at least create
> a ticket to remember about that on March, but he still thinks it to be
> unnecessary. Perhaps he is right, if the complaints to come from the
> stable, not the SVN, and the issues will continue to appear in this list
> so as to keep us aware of the problem. However, if the new version does
> not work as expected there will be six more months until the eventual
> needed fix be introduced, and go into the following release.
>
> []s,
> Fernando

I've implemented my idea at Cross-LFS; you can see the page here - 
http://cross-lfs.org/view/svn/x86/prologue/hostreqs.html - I've also 
added slightly more explicit descriptive text.

Also, I find it funny how often the sh -> dash issue is pointed out as 
an issue, yet to my knowledge it hasn't caused any actual problems for 
some time now. Makes me wonder if it's only there to verify whether 
users are really reading the book, rather than for any real, practical 
purposes.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] sed 'no such file' problem - LFS 7.1, Section 6.13, Binutils-2.22

[Chris Staub]

On 01/06/2013 06:15 PM, Ahmed ELtayeb wrote:
> just
>> cd binutils-2.22
> and everything will be fine
>
>
>
> Valentine Michael Smith  gmail.com> writes:
>
>>
>>
>> I removed the outdated standards.info file, and then tried the sed 
>> command:sed
> -i.bak '/^INFO/s/standards.info //' etc/Makefile.inI got back:sed: can't read
> etc/Makefile.in: No such file or directoryI checked, and sure enough there was
> no such file.It seems someone else on the mailing list had the same problem 
> back
> in March, 2012 with LFS 6.8, but I can't find an answer to it.Thanks.


Re-read page 5.3 carefully.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Boot LFS from usb

[Chris Staub]

On 01/07/2013 07:18 PM, Baho Utot wrote:
> I am trying to put LFS on a thumbdrive to use a rescue/fixit system.
>
> When I boot it can not find the usb drive and filesystem.  I have the
> correct filesystem built into the kernel, not as a module
>
> I am miising something in my kernel config.
> This system boots fine from a hard drive, just won't boot when I put it
> on a thumb drive
>
> Any one known what needs to be set in the kernel to allow booting from
> thumb drive?

You don't need to set anything in the kernel - it just takes some time 
for it all to be loaded from USB, so you only need a delay. Add 
"rootdelay=10" to the kernel line in your bootloader config (feel free 
to adjust the number, could try lower if you want to see if it will work 
with less delay, or higher if needed).
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Issues with compiling GCC-4.8.1 (LFS 7.4)

[Chris Staub]

On 11/02/2013 08:14 AM, Bernard Hurley wrote:
> Hi there!
>
> I have been trying to compile GCC-4.8.1 in section 5.5 of LFS 7.4.  The
> first problem was that the compiler didn't seem to understand (a
> parameter to configure on p.37):
>
>--disable-libstdc++-v3
>
> In case anyone else has the problem, I found after running "configure
> --help" that the solution seems to be to replace it with:
>
>--disable-libstdcxx

--disable-libstdc++-v3 works just fine, so you must have mistyped something.
>
> The second problem is a bit more intractible.  Having made the apbove
> adjustment to "configure", when I run "make" it bombs out with:
>
>  snip ===
> configure: creating ./config.status
> config.status: creating Makefile
> config.status: creating auto-target.h
> config.status: linking ../.././libgcc/enable-execute-stack-empty.c to 
> enable-execute-stack.c
> config.status: linking ../.././libgcc/unwind-generic.h to unwind.h
> config.status: linking ../.././libgcc/config/i386/linux-unwind.h to 
> md-unwind-support.h
> config.status: linking ../.././libgcc/config/i386/sfp-machine.h to 
> sfp-machine.h
> config.status: linking ../.././libgcc/gthr-single.h to gthr-default.h
> config.status: executing default commands
> make[2]: Entering directory 
> `/mnt/lfs/gcc-build/gcc-4.8.1/x86_64-lfs_7_4-linux-gnu/libgcc'
> Makefile:162: ../.././gcc/libgcc.mvars: No such file or directory
> make[2]: *** No rule to make target `../.././gcc/libgcc.mvars'.  Stop.
> make[2]: Leaving directory 
> `/mnt/lfs/gcc-build/gcc-4.8.1/x86_64-lfs_7_4-linux-gnu/libgcc'
> make[1]: *** [all-target-libgcc] Error 2
> make[1]: Leaving directory `/mnt/lfs/gcc-build/gcc-4.8.1'
> make: *** [all] Error 2
>  snip ===

One problem I'm seeing in this output is that you are running this from 
$LFS/gcc-build, and have apparently unpacked GCC's source inside that 
dir, which is not what you're supposed to be doing. Go back and read 
page 5.3 very carefully.

>
> I have the following environment variables set:
>
> LFS=/mnt/lfs
> LFS_TGT=x86_64-lfs_7_4-linux-gnu
> PATH=/tools/bin:/bin:/usr/bin

Apparently you've modified $LFS_TGT from what the book says to use. 
Granted, this particular change should not be a problem, but it does 
indicate that you've deviated from the book's instructions. In what 
other ways have you done things differently from the book?

>There appears to be somthing wrong with the "sed" command on page 37:

 >   sed -i '/k prot/agcc_cv_libc_provides_ssp=yes' gcc/configure

 >as the output from 'grep "k prot" gcc/configure' (before running "sed")
 >is:

 >  # Test for stack protector support in target C library.
 ># all versions of Bionic support stack protector

There is nothing wrong with that sed command - you are just 
misunderstanding what it actually does. Try changing the "-i" to 
"-i.bak" and then doing a "diff -au gcc/configure{.bak,}" to see the 
differences.

> Another thing that seems strange is the configure option (again on page
> 37):
>
>--with-mpfr-lib=$(pwd)/mpfr/src/.libs
>
> There is no file "mpfr/src/.libs", in fact there is nothing called
> ".libs" in the whole of mpfr's source tree.

Are you saying that you are getting an error of some kind about this? 
That directory should be created during compile, so it shouldn't matter 
that it doesn't exist when you unpack the source.

> If you've read so far you are truly one of my heros!  I hope to be able
> to contribute to the community myself once I have built LFS.
>
> Kind regards,
>
> Bernard.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Issues with compiling GCC-4.8.1 (LFS 7.4)

[Chris Staub]

On 11/02/2013 12:52 PM, Chris Staub wrote:
>>
>> The second problem is a bit more intractible.  Having made the apbove
>> adjustment to "configure", when I run "make" it bombs out with:
>>
>>  snip ===
>> configure: creating ./config.status
>> config.status: creating Makefile
>> config.status: creating auto-target.h
>> config.status: linking ../.././libgcc/enable-execute-stack-empty.c to 
>> enable-execute-stack.c
>> config.status: linking ../.././libgcc/unwind-generic.h to unwind.h
>> config.status: linking ../.././libgcc/config/i386/linux-unwind.h to 
>> md-unwind-support.h
>> config.status: linking ../.././libgcc/config/i386/sfp-machine.h to 
>> sfp-machine.h
>> config.status: linking ../.././libgcc/gthr-single.h to gthr-default.h
>> config.status: executing default commands
>> make[2]: Entering directory 
>> `/mnt/lfs/gcc-build/gcc-4.8.1/x86_64-lfs_7_4-linux-gnu/libgcc'
>> Makefile:162: ../.././gcc/libgcc.mvars: No such file or directory
>> make[2]: *** No rule to make target `../.././gcc/libgcc.mvars'.  Stop.
>> make[2]: Leaving directory 
>> `/mnt/lfs/gcc-build/gcc-4.8.1/x86_64-lfs_7_4-linux-gnu/libgcc'
>> make[1]: *** [all-target-libgcc] Error 2
>> make[1]: Leaving directory `/mnt/lfs/gcc-build/gcc-4.8.1'
>> make: *** [all] Error 2
>>  snip ===
>
> One problem I'm seeing in this output is that you are running this from
> $LFS/gcc-build, and have apparently unpacked GCC's source inside that
> dir, which is not what you're supposed to be doing. Go back and read
> page 5.3 very carefully.

Further, if you've managed to create $LFS/gcc-build, it means you either 
chowned $LFS to the lfs user, or temporarily became root to create 
$LFS/gcc-build, neither of which the LFS book says to do. In general, if 
you get some kind of permissions error, you should stop and double-check 
that you're doing what you're supposed to, rather than forcing root. If 
the book is correctly followed, you should be able to do all of Chapter 
5 entirely as the lfs user. Also, if you did chown $LFS, you should 
chown it back to root.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] LFS 7.4 / 5.5 GCC Pass 1 Question

[Chris Staub]

On 11/24/2013 04:01 PM, Kenan Regez wrote:
> Hi,
>
> It is my first try to build LFS im really excited about the project but
> i have got a question about the step GCC 4.8.1.
>
> I run into the same error many times when i run "make". I followed every
> step in the book, i untared the gmp, mpfr and mpc into my gcc-4.8.1 dir,
> as written in the book i renamed the folders.
>
> Now after spending some hours to find the problem i did this another way
> which worked but im unsure if it legit to use the way i did.
>
> I got allways this error:
>
> |checkingfor  MPFR...  no
> configure:  error:  libmpfr not found or uses a different ABI.
> make[1]:  ***  [configure-mpc]  Error  1|

This usually means that mpfr was not unpacked where it should be. Please 
paste, directly from your command history, all the exact commands you 
typed which lead to this error, starting with unpacking the GCC tarball.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] lfs 7.4 / Section 6 - Building the LFS System as sudo?

[Chris Staub]

On 11/29/2013 01:56 PM, frozen tuesday wrote:
> Hello all --
>
> Thanks to the help I have received here or elsewhere, I have made it as
> far as section III in the online instructions: "Building the LFS System".
>
> I am using Linux Mint 15 as my host computer which only allows me to
> "sudo" before commands. I cannot log in as root (to my knowledge). The
> instructions at the end of section 5 state:
>
> "The commands in the remainder of this book must be performed while
> logged in as user |root| and no longer as user |lfs|. Also, double check
> that |$LFS| is set in |root|'s environment."
>
> Is it sufficient to sudo before every command in the rest of the book,
> or do I need to find a way to log in as the user "root"?
>
> Thanks.

You stay root anyway as long as you're in the chroot.

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] lfs 7.4 / Section 6 - Building the LFS System as sudo?

[Chris Staub]

On 12/01/2013 06:27 PM, frozen tuesday wrote:
>  >> Is it sufficient to sudo before every command in the rest of the book,
>  >> or do I need to find a way to log in as the user "root"?
>  > >
>  > > Thanks.
>
>  > You stay root anyway as long as you're in the chroot.
>
> I see the chroot is coming up in 6.4 but before that here are some
> commands that need to be entered in the 6.2 section. Do I enter those as
> sudo or as lfs?
>
> Thanks.

The book tells you. See the "Changing Ownership" page again.

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] GCC-4.8.1 Pass1 C++

[Chris Staub]

On 01/11/14 03:35, William Darryl Jackson wrote:
> Greetings:
>
> Having rebuilt from scratch, and being certain all steps were followed
> accurately - in the proper folder and environment; GCC 'make' is still
> giving me the following error:
>
> configure: error: C++ compiler missing or inoperational
> make[1]: *** [configure-libcpp] Error 1
> make[1]: Leaving directory `/mnt/lfs/sources/gcc-build'
> make: *** [all] Error 2
>
> Again, Binutils installed correctly.
>
> Any assistance would be greatly appreciated.
>
> William
>
Paste the contents of libcpp/config.log.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] folder permissions

[Chris Staub]

On 01/11/14 14:25, William Darryl Jackson wrote:
> Yes,
>
> mount
>
> /dev/sdb2 on /media/lfs type ext4
> (rw,nosuid,nodev,relatime,user_xattr,barrier=1,data=ordered,uhelper=udisks)
>
> The echo $LFS, is constantly slipping in and out of /mnt/lfs. I have to
> keep doing the 'export'. Right now it returns nothing. The only time it
> did not slip-out, unmount itself was when I kept the same terminal open
> from start to finish. However, close that terminal - reboot, changes
> everything. I am thinking this might be Debian, if so going to internal
> drive would not help.
>
> Thanks.
>
> William

Your output says there is something mounted on /media/lfs, yet you keep 
saying you are assigning LFS=/mnt/lfs.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Chapter 5.7 glibc-2.1{7,8}

[Chris Staub]

On 01/11/14 14:38, Frans de Boer wrote:
> Dear reader,
>
> I am having problems lately making the tool chain on my x86_64 machine.
> I finally traced it back to chapter 5.7, where glibc aborts with the
> next messages:
>
> Makerules:755: *** mixed implicit and normal rules.  Stop.
> make[1]: Leaving directory `/mnt/lfs/sources/glibc-2.17'
> make: *** [all] Error 2
>
> These messages are only on my 64-bit machine. Using a i686 machine
> yields no problem and I can build LFS. I use standard scripts which are
> an exact copy of the book. The same scripts only fail on my AMD phenom
> II machine.

They can't be an exact copy because the book does not say to create 
glibc-2.17/build. Is there anything else you have changed?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] folder permissions

[Chris Staub]

On 01/11/14 14:47, William Darryl Jackson wrote:
> On 01/11/2014 02:33 PM, Chris Staub wrote:
>> On 01/11/14 14:25, William Darryl Jackson wrote:
>>> Yes,
>>>
>> Your output says there is something mounted on /media/lfs, yet you keep
>> saying you are assigning LFS=/mnt/lfs.
> I was trying to follow the instructions precisely. I agree this seems to
> be the cause of the problem. This is an external drive that auto mounts
> to the (user) media folder, and (root) dev folder. The only reason it
> says media/lfs is because I added a label in gparted. Debian is using
> uuid identifiers - which is causing me problems. Would it help if I
> replaced LFS=/mnt/lfs with something else? Also, I have 2 partitions on
> that external drive. They both mount showing 500GB, when the drive is
> only 500GB.
>
> Thanks.
>
Sorry, but if you really need to ask this, LFS is just too advanced for 
you. I suggest going here - 
http://www.linuxfromscratch.org/lfs/view/7.4/prologue/prerequisites.html 
- for starters, and learning about how to actually use Linux, especially 
on the command line, before attempting something like LFS.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Two suspected bugs

[Chris Staub]

On 01/16/14 13:00, parmenides wrote:
> Hi,
>
> I am build the LFS 7.4 following the book, and find two suspected bugs:
>
> (2) In chapter 5, "5.7. Glibc-2.18"
> I got the following warning:
>
> configure: WARNING:
> *** These auxiliary programs are missing or incompatible versions: autoconf
> *** some features will be disabled.
> *** Check the INSTALL file for required versions.
>
> I seems that there should be some requirements of autoconf.

No, Autoconf is not needed. That is just warning, not an error.

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] Gcc-4.8.1 Section 5.5.1 - libmpfr

[Chris Staub]

On 01/16/14 18:07, William Darryl Jackson wrote:
> On 01/16/2014 04:23 PM, Pierre Labastie wrote:
>> Le 16/01/2014 22:15, William Darryl Jackson a écrit :
> You are correct - Pierre, I came to that same conclusion. So I very
> carefully repeated the steps on page 36 to ensure those folders were
> being built correctly (owner/group=lfs) then (delete contents of
> gcc-build; reconfigure, and make) -No change; this error must be
> pointing to something else with my system.
>
> My Internet searches indicate that others have had similar problems, but
> I saw no resolution. What if any is the host interaction at this point?
> Gotta find somewhere else to look.
>
> Thanks,
>
> William
>
Type "history" as the lfs user, and paste every command you have run for 
building GCC, starting with unpacking its source tarball.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: [lfs-support] CLFS

[Chris Staub]

On 02/19/14 05:46, loki wrote:
> Heya,
>
> just wanted to ask what happened to CLFS?
>
> On the address http://www.cross-lfs.org/ I'm getting a Domain for Sale.
>
> Regards,
> Daniel
>
>
Nothing happened to it, it's where it's always been, at trac.cross-lfs.org.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: jhalfs?

[Chris Staub]

Galaxy Travel wrote:
> How do you install this?
> 

Unpack the tarball and read the README.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: jhalfs?

[Chris Staub]

Galaxy Travel wrote:
>> Unpack the tarball and read the README.
>> -- 
>> http://linuxfromscratch.org/mailman/listinfo/lfs-support
>> FAQ: http://www.linuxfromscratch.org/lfs/faq.html
>> Unsubscribe: See the above information page
>>
> 
> I did that it says make menuconfig which I did then what? it dont tell you?
> 

Exactly what part of menuconfig are you having difficulty with?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: jhalfs?

[Chris Staub]

Galaxy Travel wrote:
> - Original Message - 
> From: "Chris Staub" <[EMAIL PROTECTED]>
> To: "LFS Support List" 
> Sent: Saturday, February 10, 2007 12:18 PM
> Subject: Re: jhalfs?
> 
> 
>> Galaxy Travel wrote:
>>>> Unpack the tarball and read the README.
>>>> -- 
>>>> http://linuxfromscratch.org/mailman/listinfo/lfs-support
>>>> FAQ: http://www.linuxfromscratch.org/lfs/faq.html
>>>> Unsubscribe: See the above information page
>>>>
>>> I did that it says make menuconfig which I did then what? it dont tell 
>>> you?
>>>
>> Exactly what part of menuconfig are you having difficulty with?
>> -- 
>> http://linuxfromscratch.org/mailman/listinfo/lfs-support
>> FAQ: http://www.linuxfromscratch.org/lfs/faq.html
>> Unsubscribe: See the above information page
>>
> Maybe im missing something but what do I do after make menuconfig? 
> 
> 

Immediately after you exit menuconfig, jhalfs should run (or at least 
tell you what to do), as mentioned in the README.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: NCURSES 5.6 BUG

[Chris Staub]

Ag. Hatzimanikas wrote:
> On Mon, Feb 12, at 08:30 [EMAIL PROTECTED] wrote:
>> Can you please specify what the bug is, and perhaps provide a link to the 
>> patch?
> 
> I guess he is referring to include/MKhashsize.sh
> Particularly in the following variable.
> 
> TABSIZE=`grep -v '^[ #]' $CAPS | grep -v "^$" | grep -v "^capalias"|\
> grep -v "^infoalias" | wc -l`
> 
> The problem I think is the backticks with the square brackets.
> 
> This can be solved,either changing the variable to the form $(...)
> or just enclose the variable within double quotes "`...`"
> Even the following sed will suffice,as the grouping isn't really
> necessary.
> sed -i 's/\[ #\]/#/' include/MKhashsize.sh
> 
> This was reported upstream by Greg Schafer,but I don't know the current 
> status,
> I thought it was fixed in the last bash patch.
> 
> I saw the same problem once,in one build with bison,in the build-aux/ylwrap.
> 

This does appear to be the case...I built Bash 3.2 without the upstream 
patch, and used it to try compiling ncurses...it does indeed fail with 
that error. If I attempt to compile ncurses using /bin/bash, which was 
built with the patch, it works fine, so it seems the problem has been 
fixed with the upstream patch. I guess "Galaxy Travel" simply forgot to 
apply the patch.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Udev Errors?

[Chris Staub]

Galaxy Travel wrote:
> - Original Message - 
> From: "Dan Nicholson" <[EMAIL PROTECTED]>
> To: "LFS Support List" 
> Sent: Wednesday, February 14, 2007 5:17 PM
> Subject: Re: Udev Errors?
> 
> 
>> On 2/14/07, Galaxy Travel <[EMAIL PROTECTED]> wrote:
>>> Im having  a couple errors when I reboot the system, now my question is
>>> where are the logs kept for udev so I can show you what errors?
>> They should be in syslog (/var/log/sys.log) if the errors occurred
>> after syslogd starts.
>>
>> --
>> Dan
>> -- 
>> http://linuxfromscratch.org/mailman/listinfo/lfs-support
>> FAQ: http://www.linuxfromscratch.org/lfs/faq.html
>> Unsubscribe: See the above information page
> 
> Wasnt in sys.log but i managed to write it down:
> 
> Couple Errors:
> 
> udevd-event[1154]: run program: exec of program '/lib/udev/path_id' failed
> 
> theres a couple of them with different event then also have usb_id failed.
> 
> Whats the problem? 

Check to see if /lib/udev/path_id and /lib/udev/usb_id exist. If not, 
you probably left them out when you built udev.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: first pass GCC bootstrap problem

[Chris Staub]

Nick Bowes wrote:
> - Original Message - 
> From: "Dan Nicholson" <[EMAIL PROTECTED]>
> To: "LFS Support List" 
> Sent: Thursday, February 22, 2007 7:09 PM
> Subject: Re: first pass GCC bootstrap problem
> 
> 
>> If it's your first time following the book, chances are that you did
>> get confused (a lot of people do, including me). The idea is that you
>> have the /sources directory with all the tarballs and patches. When
>> you start a section, the commands on the page expect that you've
>> unpacked the associated tarball and entered the source tree. The
>> commands shouldn't need to be adjusted if you're in the right
>> location.
>>
>> I should also mention that using /sources isn't strictly necessary.
>> But, if it's your first time, I would suggest just following the book
>> since it is known to work. You can always tweak things to your liking
>> the next time around.
> 
> Excellent! did exactly as you said and it's solved the problem. I am 
> however having a hard time deleting the source and build directories. 
> If  I do "rmdir -v --ignore-fail-on-non-empty gcc-build/", it prints 
> "rmdir: removing directory, gcc-build/" but does nothing and returns to 
> the promt immediatly. Same for the binutils-build. Is there something 
> i'm missing or should this work?
> 
> Thanks for your help Dan 
> 

That's because rmdir only removes empty directories, so it won't remove 
gcc-build if it's not empty...

--ignore-fail-on-non-empty
   ignore each failure that is solely because a directory
   is non-empty
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Coreutils - LFS 6.2

[Chris Staub]

Bartlomiej P. Lyson wrote:
> Hello,
> I've got a problem with running the test coreutils.
> 
> My lfs version is 6.2.
> 
> an example: 
> [EMAIL PROTECTED]:~# /usr/sbin/useradd -m test
> 
> [EMAIL PROTECTED]:~# cat /etc/passwd | grep test
>   test:x:1001:1002::/home/test:/bin/bash
> 
> [EMAIL PROTECTED]:~# cat /etc/group | grep test
> test:x:1002:
> 
> [EMAIL PROTECTED]:~# su - test
> id: cannot find name for group ID 1002
> id: cannot find name for user ID 1001
> 
> 
> Logs come from Coreutils-6.7 but before I was testing
> on voresion from LFS 6.2, exactly 5.96 and was same
> problem.
> 
> part logs from test run:
> ==
> 
> Maybe someone had the same problem or maybe someone
> knows what the problem is ?
> 
> Regards
> --
> Bartlomiej P. Lyson
> 
> 

Did the method for running the coreutils testsuite as described in the 
book, with the temporary users and groups it says to create, not work 
for you?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Coreutils - LFS 6.2

[Chris Staub]

Bartlomiej P. Lyson wrote:
> Hello,
> I've got a problem with running the test coreutils.
> 
> My lfs version is 6.2.
> 
> an example: 
> [EMAIL PROTECTED]:~# /usr/sbin/useradd -m test
> 
> [EMAIL PROTECTED]:~# cat /etc/passwd | grep test
>   test:x:1001:1002::/home/test:/bin/bash
> 
> [EMAIL PROTECTED]:~# cat /etc/group | grep test
> test:x:1002:
> 
> [EMAIL PROTECTED]:~# su - test
> id: cannot find name for group ID 1002
> id: cannot find name for user ID 1001
> 
> 
> Logs come from Coreutils-6.7 but before I was testing
> on voresion from LFS 6.2, exactly 5.96 and was same
> problem.
> 
Wait a minute...you shouldn't have useradd at this point in the 
build...at exactly what part of the book are you?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: chroot chapter 6.6

[Chris Staub]

nick bowes wrote:
> Hi,
> I can't figure out what should happen in ch6.8.  the first command your 
> asked to run is "make install" but where should I run it from? should I be 
> running it as root in the previously untared man-pages-2.42 dir?
> Sorry I know this should be obvious but as you will have guessed this is all 
> very new to me.
> 
> Thanks
> Nick 
> 

I believe you are referring to the "Man Pages" section...that is the 
man-pages package you are installing (as noted on the title of the page).
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: Retrieving Packages from LiveCD after chroot

[Chris Staub]

John Rodenbiker wrote:
> Actually, it says "Downloaded packages and patches will need to be 
> stored somewhere that is conveniently available throughout the entire 
> build." It goes on to say that a working directory will also be 
> required and that $LFS/sources can work for both. The text does not say 
> it is necessary
> 
> As I said, I naively thought that the sources on the Live CD _would be_ 
> conveniently available throughout the entire build. To avoid a long 
> initial copy from CD and using a ton of disk space unnecessarily, I 
> thought it made sense to only cp and unpack tarballs and patches as 
> they were required and then delete them after a successful install.
> 
> I humbly suggest that the text in 3.1 be edited to remove ambiguity and 
> explicitly say that all sources need to be put in $LFS/sources from the 
> start.
> 

Actually, this would probably be a good idea, but (IMO) not for this 
reason. Another problem is that you would likely run into problems if 
you compile in /tools (as some people seem to interpret the 
instructions) as Glibc apparently does something weird when it sees an 
"include" directory. We have already changed the instructions in 
Cross-LFS to say that you are expected to use $CLFS/sources and that 
doing otherwise may have unexpected results.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: 7.5. Configuring the setclock Script - no /etc/sysconfig

[Chris Staub]

Kai Ulrich wrote:
> Hey,
> 
> in chapter I supose to create a /etc/sysconfig/clock file.
> 
> by executing the comand
> 
> cat > /etc/sysconfig/clock << "EOF"
> # Begin /etc/sysconfig/clock
> 
> UTC=1
> 
> # End /etc/sysconfig/clock
> EOF
> 
> i get the message,
> 
> /etc/sysconfig/clock: No such file or directory
> 
> and true there is no foulder /etc/sysconfig
> 
> Went something wrong or do I just have to create this folder ?
> 
> regads
> kai

That directory should be installed by the installation of 
lfs-bootscripts. Go back a few pages in the book and make sure you've 
installed the lfs-bootscripts package.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: GCC make install problem

[Chris Staub]

Bob Newell wrote:
> I am trying to build my LFS 6.2, and I am currently in
> section 5.4.1 (building GCC).  So far I havent
> deviated from the book other than the mount point of
> my LFS sources. When I got to the "make install"
> command, here is what I get
> 
> [EMAIL PROTECTED]:/media/hdc2/sources/gcc-build$ make
> install
> /bin/sh ../gcc-4.0.3/mkinstalldirs /tool /tool
> mkdir -p -- /tool /tool
> mkdir: cannot create directory `/tool': Permission
> denied
> mkdir: cannot create directory `/tool': Permission
> denied
> make: *** [installdirs] Error 1
> 
> I can't find anything in the FAQ or mailing list
> archives.  Thanks in advance for your help.
> 
> Bob Newell

According to the error message it is trying to install to "/tool", when 
it should be "/tools". Assuming this is an accurate copy-and-paste of 
the error, then you must have specified the wrong directory with 
./configure. You'll need to remove the GCC source and build directories 
and recompile.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Re: 6.9.1 Glibc-2.3.6 problems...

[Chris Staub]

Peter Santimauro wrote:
> After mkdir -v ../glibc-build  and cd ../glibc-build, I tryed to run
> the configure script.  I  get the following as and output:
> ...
> the config.log  says:
> 
> ...
> configure:2631: checking for C compiler version
> configure:2634: gcc --version  &5
> ../glibc-2.3.6/configure: line 2635: /bin/gcc: No such file or directory
> configure:2637: $? =127
> configure:2639: gcc -v  &5
> ../glibc-2.3.6/configure: line 2640: /bin/gcc: No such file or directory
> configure:2642: $? =127
> configure:2644: gcc -V  &5
> ../glibc-2.3.6/configure: line 2645: /bin/gcc: No such file or directory
> configure:2647: $? =127
> configure:2651: checking for suffix of object files
> configure:2672: ggc -c conftest.c >&5
> ../glibc-2.3.6/configure: line 2673: /bin/gcc: No such file or directory
> configure:2675: $? =127
> configure: failed program was:
> | /* confdefs.h.  */
> |
> | ...confdefs contents ...
> |
> |
> configure: 2689: error: cannot compute suffix of object files: cannot compile
> 
> 
> 
> any ideas why this is happening?
> i tryed to run gcc but it isnt there.  this is the second time around
> for me trying to get this to work. Last time I got this error and
> restarted, I really don't want to have to do that again :(
> 
> thank
> pete

According to config.log it's looking for /bin/gcc, which should not 
exist. What is the output of "ls -l /bin/gcc" and "ldd /bin/gcc"?
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page